Jlfi Northern Territory Naturalist FIELD NATURALISTS’I CLUBJnc. Number 27 October 2016 The Journal of the NT Field Naturalists’ Club NORTHERN TERRITORY FIELD NATURALISTS’ CLUB Inc. Founded 1977 Club officers for 2015/16 President: Richard Willan Secretary: Julie Wilson Treasurer: Ilona Barrand Northern Territory Naturalist editors Richard Willan (Chief Editor) Sean Belairs Peter Kyne Sue Dibbs (Production Editor) ISSN 0155-4093 2016 Northern Territory Field Naturalists’ Club Inc. Previous issue (Number 26) published 9 June 2015 The objectives of the Northern Territory Field Naturalists’ Club are to promote the study and conservation of the flora and fauna of the Northern Territory. The Club holds monthly meetings and field excursions. Meetings are held in room Red 1.03 at the Charles Darwin University Casuarina Campus, Darwin, at 7.45 pm on the second Wednesday of each month. All members receive the monthly newsletter Nature Territory and the journal Northern Territory Naturalist. For information on membership, club activities and publications, please write to: Northern Territory Field Naturalists’ Club Inc. PO Box 39565, Winnellie, NT 0821 or visit our website: Guidelines for authors may be downloaded from this website. Front Cover: Mangrove leaf slugs newly recorded from the Northern Territory: Elysia leucolegnote (above) and E. bangtawaensis (below). (Adam Bourke) Northern Territory Naturalist Contents New insights into Holocene economies and environments of Central East Timor: Analysis of the molluscan assemblage at the rockshelter site of Hatu Sour Sally Brockwell, Sue O’Connor, Mirani Ulster and Richard C. Willan 2 The pathogen Myrtle Rust (Puccinia psidii) in the Northern Territory: First detection, new host and potential impacts John 0. Westaway 13 Asystasiagangetica subsp. micrantha , a new record of an exotic plant in the Northern Territory John 0. Westaway, Lesley Alford, Greg Chandler and Michael Schmid 29 Seed viability of native grasses is important when revegetating native wildlife habitat Sean M. ReHairs and Melina ]. Caswell 36 Nest site fidelity of Flatback Turdes (Nata/or depressus) on Bare Sand Island, Northern Territory, Australia Natalie Bannister, John Holland and Trisia I 'arrelly 47 Records of waterbirds and other water-associated birds from the 2014/15 migratory season in the Darwin region Amanda UHeyman 54 Fluctuations in use of urban roost and foraging sites in Darwin by Pied Herons ( Ardeapicata) John Rawsthorne 60 Evidence of rock kangaroo seed dispersal via faecal seed storage in a tropical monsoon community Christopher L Marline, Alexandra /. Bold, Elizabeth A. Capa/di, Gemma E. Uonheart and Ingrid E. Jordon-Thaden 68 Report of the presence of 1 \apahtrema synorchis and /1. postorchis (Digcnea: Spirorchiidae) in marine turtles (Reptilia: Cheloniidae) in Northern Territory waters Diane P. Barton, Phoebe A. Chapman and Rachel A. Groom 78 Coral communities in extreme environmental conditions in the Northern Territory, Australia Lawrance VC. I 'erns 84 First record of two mangrove leaf slugs, Efysia leucolegnote and E. bangtawaensis (Sacoglossa: Plakobranchidae), in mangrove forests in the Northern Territory Adam J. Bourke, Carmen Walker and Richard C. Willan 97 Field identification of the Platanndex mangrove slugs (Mollusca: Gastropoda: Onchidiidae) of Darwin Harbour Adam J. Bourke 102 Captain King’s lost weevil - alive and well in the Northern Territory? Stefanie K. Oberprie/er, Debbie Jennings and Rolf G. Oberprie/er 106 Rediscovery of the Spinifex Sand-skipper ( Proeidosapofysema) in the Darwin area. Northern Territory Michael F. Braby and John O. Westaway 121 Book Review John Dengate 126 Advice to authors Inside back cover Northern Terri tor)' Naturalist (2016) 27: 2—12 Research Article New insights into Holocene economies and environments of Central East Timor: Analysis of the molluscan assemblage at the rockshelter site of Hatu Sour Sally Brockwell 1 , Sue O’Connor 1 , Mirani Litster 1 and Richard C. Willan 2 ' Department of Archaeology and Natural History, School of Culture and History and Language, College of Asia and Pacific, Australian National University, Canberra, ACT 0200, Australia Email: sally.brockwcU@anu.edu.au 2 Museum and Art Gallery of the Northern Territory, GPO Box 4646, Darwin, NT 0801, Australia Abstract In the central region of East Timor (the proper name for this nation being Timor- Leste) litdc is known of prehistoric economies beyond 2000 years ago, most previous archaeological studies having been concentrated around the Baucau plateau and eastern end of the island. The village of Lalcia on the Laleia River is located 20 km cast of the main district town of Manatuto on the central northern coast. Recent excavations at the nearby rockshelter site of Hatu Sour have revealed a deep archaeological sequence that dates from approx. 11,000 years ago until the recent past. This paper examines the shellfish (i.e., molluscan) assemblage from the excavation at Hatu Sour for what it can reveal about prehistoric economies and the environment of this strategic region throughout the Holocene. Introduction This study was undertaken as part of an Australian Research Council Discovery Project investigating cultural and environmental shifts in Holocene Hast Timor. This investigation involved a programme of excavation at various archaeological sites to document cultural and environmental histories. One of the key cultural remains in these sites is shellfish (Mollusca), which provide evidence not only of what dietary preferences existed but also what shellfish were locally available and therefore what niches were being exploited. This is also an excellent indicator of past environments, as shellfish are extremely sensitive to changes in their environment. Up until 2010 most of our research had been undertaken at the far eastern end of the island of Timor in the Lautcm District. One of our collaborators, the Secretariat of Culture, East Timor Government, was keen for us to investigate other areas of East Timor. In 2010, we located several prospective- sites in the Manatuto District on the central northern coast. Here we discuss the results, Molluscs in Hatu Sour rocksheher Northern Territory Naturalist (2016) 27 3 focusing on shellfish remains, of the excavation at Hatu Sour rocksheher and how they help us to interpret regional prehistoric economies and palaeo-environments. In the central region of East Timor little is known of prehistoric economies beyond the last 2000 years, most previous archaeological studies being concentrated around the Baucau plateau and the eastern end of the island (Almeida & Zbyszcwski 1967; Cilover 1986; O’Connor, Spriggs & Veth 2002; O’Connor 2003; O’Connor & Veth 2005; Selimiotis 2006; O’Connor 2007; Oliveira 2008, 2010; O’Connor 2010; O’Connor et al. 2010, Oliveira 2010; O’Connor el al 2011, Reepmeyer el al 2011; O’Connor el al. 2012). Recent excavations at the rocksheher site of Hatu Sour on the central northern coast have revealed a deep archaeological sequence that dates from approx. 11,000 years until the recent past. The site contains a large shellfish assemblage, vertebrate faunal remains and stone artefacts, and was occupied during a period of dramatic climatic and environmental change that profoundly affected the subsistence and settlement patterns of coastal dwellers. It encompasses the Neolithic transition dating from some 3500 years ago in Timor, which originally brought pottery and possibly the domestic dog, and later subsistence agriculture and other domestic animals (O’Connor 2006). It also covers the contact period from about 1000 years ago when outside influences from Chinese and later Makassar traders to Dutch and Portuguese colonisation deeply affected the indigenous culture and economy (O’Connor el al. 2012). Fig 1. Map of study area. (CartoGIS). 4 Northern Territory Naturalist (2016) 27 Brockwell et al Above left — Fig. 2. Approaching I latu Sour rockshelter. (Sue O’Connor) Above right - Fig. 3. I latu Sour rockshelter and excavation. (Sally Brockwell) Study Area East Timor (the proper name for this nation being Timor-Leste) is located 400 km north¬ west of Australia in the Timor Sea, and 8 degrees south of the Equator. It shares a land border with West Timor (the proper name for this part of the nation of Indonesia being Timor Barat). Geologically, Timor is an aggressively uplifted coral limestone island. The northern and southern coasts are divided by a steep mountain range rising to 3000 m. The climate is dry tropical with a long dry season from May to November and a shorter wet season from December to April. The study was undertaken on the central northern coast, east of the capital, Dili (Fig. 1 inset). The Hatu Sour rockshelter is located on the Laleia River, 20 km east of the main district town of Manatuto and adjacent to the village of Laleia (Fig. 1). The rockshelter (Figs 2, 3) is about 7 m by 5 m and is located in a limestone outcrop about 1 km west of the village of Laleia. Today Hatu Sour is 4 km south of the northern coastline, which drops steeply away to the continental shelf (O’Connor 2007: 530). There appears to have been Holocene infill within the embayment as indicated in Fig. 1, where the dark green marks represent recent deposition within the I .aleia River estuary. Previous investigations Previous investigations immediately to the east, west and south of the current study area (Spriggs, O’Connor & Veth 2003; Chao 2008; Lape & Chao 2008; Forestier & Guillard 2012) have revealed a range of archaeological sites in the region: open sites and rock shelters with cultural assemblages containing stone artefacts, invertebrate remains and, in the upper levels, earthenware and imported tradeware; rockshelters with painted art; shell middens; and fortified hilltop settlements containing ancestral graves and the remains of house sites. All these sites are dated within the last 8000 years, but most within the last 2000 years. Our surveys in the Laleia region revealed a similar range of sites, including remains of old villages with house stones and concentrations of stone artefacts, marine and estuarine shell remains, pottery and Chinese tradeware. One open site next to the Laleia Molluscs in Hatu Sour rockshelter Northern Territory Naturalist (2016) 27 5 River containing a scatter of shell and stone artefacts produced the unexpectedly early date of 9500 cal. BP for the bival ve. Anadaragrattosa (Table 1). There are also indigenous sites on hilltops fortified with stone walls. One particular hilltop setdement known as Leki Wakik has been occupied within living memory. These sites contain surface scatters of shell, stone, pottery and Chinese ceramics. One open site dated to 400 cal. BP from a surface shell sample (of the gastropod Te/escop/um le/escopium) (Table 1). This is consistent with the ages of other indigenous fortifications, both regionally and elsewhere in Timor, mostly dated to less than 1000 years old (Chao 2008; Lape & Chao 2008; O’Connor et al. 2012 ). Table 1. Radiocarbon dates from Laleia region (Hatu Sour dates after Cooling 2012: 56) Site Spit no. Lab no. Sample C14 (years BP) Cal. years BP (95.4% probability) Kampung Baru 1 Surface Wk-28440 A nadara gran os a 9007131 9840-9460 Kampung Baru 2 Surface Wk-28441 Telescoptum telescopium 798132 500-250 Hatu Sour 3 ANU#26606 charcoal 315125 460-305 Hatu Sour 12 ANU#27105 charcoal 6165140 7168-6849 Hatu Sour 35 ANU#26609 A nadara granosa 9650145 11,198-10,787 Methods We excavated a 1 m x 1 m square in arbitrary 5 cm spits using standard archaeological techniques. The deposit was sieved though a 1.5 mm wire mesh screen. Finds were initially washed and sorted on site by category (bone, stone, shell, etc). When we returned to our laboratories at the Australian National University (ANU) in Canberra, the finds were further analysed. In this case, molluscan shells were sorted by species, counted and weighed. Results were entered onto a spreadsheet according to MNI (Minimum Number of Individuals), N1SP (Number of Individual Specimens) and weight (g). Where the shells were broken, MNls were based on the part of die shell that was most commonly preserved. The same part of the shell was used consistendy for each excavation unit to estimate MNI. While this method potentially underestimates the true number of specimens it ensures that no individuals are counted more than once where pieces of one shell may be distributed over more than one excavation unit. Examples of taxa that could not be identified at ANU were sent to RCW for final determination. Results The excavation reached the limestone bedrock at 2 m, which was unexpectedly deep given the small size of the shelter (Fig. 3). The site contained large quantities of stone, shell, some bone and a few small pottery sherds on the surface. Chinese tradeware made of high-fired porcelain was also restricted to the surface. Three dates were obtained - 400 cal. BP at spit 3, 7000 cal. BP at spit 12, and 11,000 cal. BP at spit 35 (Atkinson 6 Northern Territory Naturalist (2016) 27 Brockwcll et a/. 2012: 6; Cooling 2012: 56), as detailed in Table 1. Shellfish Analysis As can be seen from Fig. 4, there was a peak in shellfish remains at spit 35 (approx. 11,000 years BP), which subsequently declined, then increased again around spits 15—12 (approx. 7000 years BP), peaked at spit 8, and declined up until the recent past. The site contained a mixture of marine, mangrove and freshwater associated molluscan species (Table 2, Fig. 5). The major marine species were Chiton sp., rock dwellers mainly found in shallow water, and Anadara gratwsa , Nerita spp. and Turbo spp., all found in the intertidal zone in shallow water. There was a large number of Turbo opcrcula, as opposed to Turbo shells themselves. The dominant mangrove species were Telescopium telescopium , Terebra/ia palustris and Geloina erosa. Telescopium telescopium is typically found in intertidal mudflats and mangrove forests (VC’illan 2013). There was only one species of freshwater or brackish taxon, Stenomelania sp. Hatu Sour Shellfish by Habitat 2-S ■VOMM/k« ■FWMM/Vl Hatu Sour Distribution of Total Shell MNI/kg is Fig. 4. Distribution of total shellfish remains. Mollusc frequency main species TOO Fig. 5. Major molluscan species (MNI). Hatu Sour Shell Weight by Habitat g/kg ■M»nn* ■ Vinfiov* ■»'*thwjt§r Fig 6. Distribution of marine (blue), mangrove (red) and freshwater (green) molluscan species (MNI/kg). Fig. 7. Distribution of molluscan species by habitat and weight. Table 2. Distribution of major molluscan species at Ham Sour (MNI). Molluscs in Hatu Sour rockshelter Northern Territory Naturalist (2016) 27 7 8 Northern Territory Naturalist (2016) 27 Brockwell et al. From 11,000 to 7000 years BP the MNI shows that there is steady foraging of both mangrove and marine molluscs with marine species dominating from spit 30 until spit 15, when mangrove species begin to take over. Stenomelania sp. was being exploited, but only in low numbers until spit 15, when its presence increased significantly. From this time post-7000 years BP, species from all habitats increased with mangrove species dominating and peaking in spit 8 (Fig. 6). There is a significant decline in all species from spit 6 onwards, although mangrove species continue to dominate. Shellfish weights show this pattern even more clearly than MNI (Fig. 7). Other archaeological evidence Turning to the evidence from the stone artefact assemblages that were analysed by Cooling (2012), there was a large peak in artefact numbers between spits 3 and 9 post 7000 BP. On the basis of dates for peaks in stone artefact deposition from other rockshelter sites in East Timor, Cooling (2012: 57, 70) cautiously assigned the peak at Hatu Sour in spits 6-8 to between 5000-3000 years BP. Other faunal remains at Hatu Sour are mosdy fishes, rodents (murids), crabs and bats (all endemic fauna), most of which appear burnt. There is also a small amount of reptile vertebrae including snake, and a few unidentified bird bones. At least three species of murid exist, one of which is quite small (probably R attus exit Ians), another larger, and the third a giant rat now extinct in Timor (two bones were found in spits 17 and 18). Most of the fish are parrotfish, a common reef fish. Exotic fauna are only found in upper levels and include remains of pig and dog (Stuart Hawkins, pers. comm.). The few small sherds of pottery and Chinese tradeware that were found were restricted to the surface of the site (Cooling 2012: 34). Discussion Ham Sour’s current distance from the shoreline is about 4 km. Due to the depth of the Ombai Strait between East Timor and the island of Alor to the north, the northern coast of the island drops away sharply (O’Connor 2007: 230). Consequently, the coastline would have been more or less stable throughout the Holocene period despite sea level rise. The distance of die rockshelter from the Lalcia River is currently 1 km. Before sea level stabilisation approx. 6000 years BP, rising seas would have flooded former embayments and river valleys, subsequendy infilling them with sediment derived from both the land and the sea (Chappell 1988; Woodroffe 1988; Woodroffe, Thom & Chappell 1993). This scenario is suggested for the Lalcia River estuary by the Google Earth image where the green shading indicates recent infill (Fig. 1). Infill would have encouraged the expansion of mangroves on the floodplains of the river (Chappell 1988; Woodroffe 1988; Woodroffe, Thom & Chappell 1993) and potentially rendered mangrove resources closer than previously. Increased sedimentation within river systems in East Timor could also be the result of increased rainfall in the mid-Holocene. A recent review (Reeves et at. 2013) from northern Australia suggests that the early to mid-Holocene was warmer and Molluscs in Hatu Sour rocksheltcr Northern Territory Naturalist (2016) 27 9 wetter than at present across tropical northern Australia with drier and more variable conditions beginning sometime after the mid-Holocene. The persistent presence of marine, mangrove and freshwater molluscan species from 11,000 years BP until recently at Hatu Sour implies that the rocksheltcr was occupied continuously throughout the Holocene and the occupants had access to all these habitats. The large number of Turbo spp. opercula (as opposed to actual Turbo spp. shell remains) may relate to the foraging strategy. Living Turbo spp. could have been processed near to their site of collection to separate the flesh from the heavy shells. However, the opercula would firmly adhere to the foot of the animal and would therefore be returned to the rocksheltcr, distorting the ratio (Szabo 2009: 197, 201). The peaks in shellfish distribution at 11,000 and 7000 years BP suggest increased regional occupation at these times with a decline in occupation between these dates. Increases in occupation can be correlated with increased productivity of the environment. The exponential increase in numbers of mangrove shellfish species from 7000 years BP can be associated with an expansion of mangroves in the river estuary as a consequence of Holocene infill, as seen in I'ig. 1, and closer proximity of mangrove resources to the rockshelter. The significant increase in the presence of Sletwmelania sp. at Hatu Sour from spits 15 to 7 could be suggestive of increased rainfall and freshwater in the environment in the mid-I lolocene that would also have allowed for an expansion of regional occupation. There is some evidence from northern Australia that this may be the case (see below). The collapse of all shellfish numbers post spit 7 suggests decreased occupation. The argument for increased occupation post 7000 BP is supported by the large peak in stone artefact numbers around spit 7 (l ; ig. 8). If Cooling (2012) is correct with her extrapolation regarding stone dates based on depth, this peak period is from 5000 to 3000 years BP. Can the decline in occupation at Hatu Sour in the late Holocene, as indicated by the decrease in numbers of shellfish and stone artefacts, be associated with drier conditions? Could it also mark a concurrent increasing reliance on subsistence farming introduced about 3500 years BP? After spit 3 dated to 400 years BP, there is a drop off of overall artefact numbers suggesting a decrease in occupation (Cooling 2012: 38). Is this reorganisation of occupation coincident with European colonisation of the region? (cf O’Connor et al. 2012). Northern Australia has the same climatic regime as Timor, being located in the wet/dry tropics. To some extent, the events recorded at Hatu Sour mirror what was happening Fig. 8. Artefact count by spit (after Cooling 2012: 38). 10 Northern Territory Naturalist (2016) 27 Brockwell et al at sites in northern Australia during the Holocene. Following the Last Glacial Maximum approx. 20,000 years ago, rising seas flooded down-cut river valleys in the early Holocene. Subsequent sedimentation led to infill and the expansion of highly productive mangroves swamps that dominated the floodplains of northern rivers. This is known as the 'big Swamp Phase’ and is dated from 7000-5000 BP in northern Australia (Woodroffc, Thom & Chappell 1985; Chappell 1988; Woodroffe 1988; Woodroffe, Mulrennan & Chappell 1993). The high productivity of these swamps is reflected archaeologically in the form of extensive estuarine shellfish middens in rocksheltcrs along the floodplains of the East Alligator River indicating widespread exploitation of mangrove environments during this period (Schrire 1982; Allen 1987, 1996; Hiscock 1999). The decline of mangrove species in Hatu Sour from spit 6 could be similar to the period approx. 5000-3000 years BP in northern Australia where further sedimentation choked off the tidal influence and restricted mangroves to the coastal fringe and river margins (Woodroffe, Thom & Chappell 1985; Chappell 1988; Woodroffe 1988; Woodroffe, Mulrennan & Chappell 1993). This was reflected archaeologically by the concurrent decline in mangrove shellfish exploitation in the rocksheltcrs of the East Alligator River (Schrire 1982; Allen 1987,1996; Hiscock 1999). At Hatu Sour, relationships between climate change, changing environments and economic strategies can be clarified with further dating and isotope analysis of the archaeological shells (Bourke et al. 2007). Conclusion There has been continuous Holocene occupation in the Lalcia region of East Timor from approx. 11,000 years BP until the recent past. Early Holocene occupation was associated with exploitation of marine and estuarine resources, with some terrestrial fauna. Increases in artefacts and mangrove shellfish after 7000 BP suggest an increase in regional occupation linked to the spread of highly productive estuarine environments. Hatu Sour shows a subsequent decrease in occupation in the late Holocene. A similar decrease in site use has been noted at other sites in East Timor after 3000 cal BP, and has often been associated with changing setdement patterns as the population is assumed to have predominantly occupied open village sites once subsistence practices changed to a farming economy. Cave sites would have continued to be used, but as hunting bivouacs rather than base camps (Glover 1986: 206—207). Hatu Sour shows further decrease in use post 400 years BP which may be due to changing settlement patterns associated with European occupation (O’Connor et al. 2012). Acknowledgements Thank you to the people of Lalcia for allowing us access to their sites. Ministry staff, Secretariat of Culture, Timor-Leste for support, Australian Research Council Discovery Project (DP0878543) and the Australian National University for funding. Thank you to Stuart Hawkins (Australian National University) for analysing the non-molluscan faunal Molluscs in Hatu Sour rockshelter Northern Territory Naturalist (2016) 27 11 assemblage. This paper was much improved by the comments of the reviewer Patricia Bourke. References Allen H. (1987) Holocene mangroves and middens in northern Australia and Southeast Asia. Bulletin of the Indo Pacific Prehistory Association 7, 1—16. Allen H. (1996) The time of the mangroves. Bulletin of the Indo Pacific Prehistory Association 15, 193-205. Almeida A. de and Zbyszewski, G. (1967) A contribuuon to the study of the prehistory of Portuguese Timor - lithic indiustries. In: Archaeology at the 11th Pacific Science Congress (cd. Solheim W.), pp. 55—67. Asian and Pacific Archaeology Series No. 1, Social Sciences Research Institute, University of Hawaii Press, Honolulu. Atkinson L. (2012) Stratigraphy and geomorphology of the Hatu Sour Rockshelter, Timor Leste. ARCH 8030. Supervised Research Project in Archaeological Science. Unpublished Report, School of Archaeology and Anthropology, Australian National University, Canberra. Bourke R, Brockwell S., Faulkner P. and Meehan B. (2007) Climate variability in the mid to late Holocene Arnhem Land region, north Australia: Archaeological archives of environmental and cultural change. In: Climate Change and Archaeology in the Pacific: Archaeology in Oceania 42 (cd. Lape P.), pp. 91—101. Chao C-Y. (2008) A microregional approach to the social dynamics in late prehistoric Manatuto, East Timor, eleventh-eighteenth century. Unpublished PhD Thesis, Department of Anthropology, University of Washington, USA. Chappell J. (1988) Geomorphological dynamics and evolution of tidal river and floodplain systems in northern Australia. In: Floodplains Research. Northern Australia: Progress and Prospects, voL 2. (eds Wade-Marshall D. and Lovcday P.), pp. 34-57. North Australian Research Unit, Darwin. Cooling S. (2012) Investigations of site use through the analysis of die lithic assemblage from the site of Hatu Sour, East Timor. ARCH 8029: Supervised Research Project in Archaeological Science. Unpublished Report, School of Archaeology and Anthropology, Australian National University, Canberra. Forestier H. and Guillaud D. (2012) A limestone outcrop as a landmark of prehistoric settlement in the Manatuto region (East Timor). In: Unearthing Southeast Asia's Past: Selected Papers from the 12th International Conference of the European Association of Southeast Asian Archaeologists vol. 1. (eds Klokke M. and Degroot V.), pp. 13—19. National University of Singapore, Singapore. Glover 1. (1986) Archaeology in Eastern Timor, 1966-67. Terra Australis 11. Department of Prehistory, Research School of Pacific Studies, Australian National University, Canberra. Hiscock P. (1999) Holocene coastal occupation of Western Arnhem Land. In: Australian Coastal Archaeology. Research Papers in .Archaeology and Natural I lis/oryNo. 31. (eds Hall J. and McNiven I.), pp. 91—103. Archaeology and Natural History Publications, Australian National University, Canberra. Lape P. and Chao C.-Y. (2008) Fortification as a human response to late Holocene climate change in East 'Timor. Archaeology in Oceania 43, 11-21. O’Connor S. (2003) Report of nine new painted rock art sites in East Timor in the context of the western Pacific region. Asian Perspectives 42(f), 96—128. O’Connor S. (2006) Unpacking the Island Southeast Asian Neolithic Cultural Package, and Finding Local Complexity. In: Uncovering Southeast Asia’s Past: Selected Papers from the 10th International Conference of the European Association of Southeast Asian Archaeologists (eds Bacus E. and Glover I.), pp. 74—87. NIAS Press, Singapore. 12 Northern Territory Naturalist (2016) 27 Brockwell ct al. O’Connor S. (2007) New evidence from East Timor contributes to our understanding of earliest modern colonisation east of the Sunda Shelf. Antiquity 81, 523—535. O’ConnorS. (2010) Continuity in shell artefact production in Holocene East Timor. In: 50 Years of Archaeology in Southeast Asia: Essays in I lonour of Ian Glover (eds Beilina B., Bacus E., Pryce T.O. and Wisseman Christie J.), pp. 219—233. River Books, Bangkok. O’Connor S. Aplin K., St Pierre E. and Feng Y.-X. (2010) Faces of the ancestors revealed: discovery and dating of a Pleistocene-age petroglyph in Lene Hara Cave, East Timor Antiquity 84, 649-665. O’Connor S., McWilliam A., Fenner J. and Brockwell S. (2012) Examining the origin of fortifications in East Timor: Social and environmental factors. Journal of Island and Coastal Archaeology 7(2), 200-218. O’Connor S., ( )no R. and Clarkson C. (2011) Pelagic fishing at 42,000 years before the present and the maritime skills of modern humans. Science 334, 1117—1121. O’Connor S., Spriggs M. and Veth P. (2002). Excavation al Lene Hara Cave establishes occupation in East Timor at least 30,000—35,000 years ago. Antiquity 76, 45—50. O’Connor S. and Veth P. (2005) Early Holocene shell fish hooks from Lene Hara Cave, East 1 imor establish complex fishing technology was in use in Island Southeast Asia five thousand years before Austroncsian settlement. Antiquity 79, 249-256. Oliveira N. (2008) Subsistence Archaeobotany: Food Production and the Agricultural Transition in East Timor. Unpublished PhD Thesis, Department of Archaeology and Natural History, College of Asia and the Pacific, Australian National University, Canberra. Oliveira N. (2010) From Bui Ccri Uato to Bui Ceri Uato Mane: A new archacobotanical assemblage from East Timor. In: 50 Years of Irchaeo/qgy in Southeast Asia: Essays in I lonour of Ian Glover (eds Beilina B., Bacus E., Pryce T.O. and Wisseman Christie I.), pp. 78-91 River Books, Bangkok. Reepmeyer C., Ot.onnor S. and Brockwell S. (2011) Long-term obsidian use in East Timor documented at Jeremalai. Archaeology in Oceania 46, 85—90. Reeves J., Bostock H., Ayliffe L., Barrows 1. el a! (2013) Palaeocnvironmcntal change in tropical Australasia over the last 30,000 years - a synthesis by the OZ-INTI MATE group. Quaternary Science Reviews 74, 97-114. Schrire C. (1982) The Alligator Rivers: Prehistory and Ecology in Western Arnhem Land. Terra Australis 7 Department of Prehistory, Research School of Pacific Studies, Australian National University, Canberra. Selimtotis H. (2006) The Core of the Matter: Core Reduction in Prehistoric East Timor. Unpublished M. Phil I hesis, Australian National University, Canberra. s P n ggs M., O’Connor S. and Veth R (2003) Vestiges of early prc-agricultural economy in the landscape of East Timor. In: Fishbones and Glittering Emblems: Southeast Asian Archaeology. (eds Karlstrom K. and Kallen A.), pp. 49-58. Museum of Far East Antiquities, Stockholm. Szabo K. (2009) Molluscan remains from Fiji. In: The Early Prehistory of Fiji. Terra Australis 31. (eds Clark G. and Anderson A.), pp. 183-211. ANU EPrcss, Canberra. VC'tllan R.C. (2013) A key to the potamidid snails (longbums, mudcrcepcrs and tree-creepers) of northern Australia, Northern Tern tor)' Naturalist 24, 68-80. Woodroffe C.D. (1988) Changing mangrove and wetland habitats over the last 8000 years, northern Austraha and Southeast Asia. In: Floodplains Research. Northern Australia: Progress and Prospects vol. _ (eds Wade-Marshall D. and Loveday, P.), pp. 1-23. NARU, Darwin. Woodroffe C., Thom B. and Chappell, j. (1985) Development of wtdespread mangrove swamps in mtd-FIolocene times. Nature 317, 711-713. Woodroffe C., Mulrennan M. and Chappell j. (1993) Estuarine infill and coastal progradation, southern van Diemen Gulf. Sedimentary Geology 83, 257-285. Northern Territory Naturalist (2016) 27: 13-28 Research Article The pathogen Myrtle Rust (Puccinia psidii) in the Northern Territory: First detection, new host and potential impacts John O. Westaway Northern Australia Quarantine Strategy, Commonwealth Department of Agriculture, 1 Pederson Road, Marrara NT 0812, Australia Email: john.wcstaway@agriculture.gov.au Abstract The plant pathogenic fungus Myrdc Rust ( Puccinia psidii) was detected within the Northern Territory on Melville Island in May 2015, five years after its arrival in New South Wales. In July the rust was found on mainland Northern Territory on the outskirts of Darwin and in September in the Darwin suburbs. Four myrtaccous plant species were found infected by the rust including the indigenous shrub L ithomyrtus retusa, which represents a novel host for P. psidii. The mode of arrival and the ecological implications of the spread of Myrdc Rust infection across Top End vegetation and plant industries are discussed. Introduction Myrtle Rust affects plants of the Myrtaceae family which includes many weLl-known natives such as eucalypts {Eucalyptus, Cotymbia), paperbarks ( Melaleuca ), bottle brush (Melaleuca, formerly Callistemon), tea tree ( Leptospermum ) and lilly pillies ( Sysygium ). Ten percent of Australia’s flora belongs to the Myrtaceae and a considerable proportion of these plants may be vulnerable to Myrtle Rust infection. Infection by Myrtle Rust typically causes distortion or loss of new growth and partial defoliation/dieback; thus it reduces photosynthetic capacity in susceptible plants and reduces reproductive capacity in some species if fruits are also infected. This fungal pathogen could have serious impacts on commercial and native plants, affecting plant nurseries, garden centres and forestry, tea tree and Australian native food industries. Myrtle Rust taxonomy and biology Rusts are plant diseases caused by fungal pathogens, specifically basidiomycetc fungi of the order Pucciniales. Puccinia psidii is an exotic fungal pathogen of a complex of closely related species referred to as Myrtle Rust, Eucalyptus Rust or Guava Rust. It was first described from a specimen collected in 1884 from Guava (Psidiumguajavd) in South America (Winter 1884 cited in Glen et al. 2007). Myrtle Rust is now regarded as native to Central and South America (Ferreira 1983 cited in Carnegie et al 2015, Glenn et al. 2007, Ramsfield et al. 2010). It has subsequently spread to the USA (Marlatt & Kimbrough 14 Northern Territory Naturalist (2016) 27 Wcstaway 1979), across the Pacific to Hawaii (Uchida et al. 2006), Japan (Kawanishi et al 2009), China (Zhuang and Wei 2011 cited in Carnegie et al 2015) and Australia. Myrtle Rust has* recendy been recorded in South Africa (Roux et al. 2013) and also Indonesia (Me Taggart; 2015). Native rusts do occur in Australia (on multiple plant families) but are rare on Myrtaceat- with only two rusts indigenous to three host plants. Knowledge of the rust and its lift* cycle arc important in understanding the impact of the organism in its environment. Rustic can exhibit complex life cycles with multiple spore types (vegetative as well as sexual) and alternative hosts. Myrtle Rust can, however, complete its life cycle on a single host, rapidly producing enormous numbers of readily dispersed infectious urediniospores vi;\ asexual means. Furthermore, Myrtle Rust sometimes produces tcliospores which cap recombine genetic material with compatible mating types, importantly yielding adaptive' variation (Makinson 2012). A characteristic of rusts that makes them formidable plant pathogens is their ability to evolve rapidly under selective pressure. When Myrtle Rust arrived in Australia it was thought to differ morphologically from the holotype of P. psidii by lacking tcliospores. Australian material was placed in the genus Uredo which produces solely urediniospores, and was described as a new species, Uredc, rangelii (Simpson et al 2006). However, tcliospores have since been found on Australian Myrtle Rust specimens and in concurrence with a lack of molecular differences U. rangelii is now synonymised as a biotype (a strain with differential physiological characteristics) of P. psidii and not recognised as a unique species (Carnegie & Cooper 2011). Results from recent molecular analysis indicate that P. psidii specimens from Australia are closely related to those from Hawaii (Machado et al 2015) and also those recently studied from Indonesia (McTaggart et al 2015). More significantly, Australian P. psidii specimens appear to be genetically uniform and not undergoing sexual recombination, suggesting that only a single predominantly asexual biotype is currently present here. Introductions of novel strains of P. psidii would however increase the likelihood of mating compatibility leading to more genetic diversity in local Myrtle Rust populations. Dispersal Unlike many fungi that can survive on dead and decaying organic matter, rusts are obligate biotrophs dependent on living host tissue for reproduction and survival. Rusts produce huge numbers of spores for wind dispersal from one host to another (Brown & Hovmoller 2002). Rusts are renowned king-distance dispersers with, for example, one race of Wheat Stripe Rust spreading from Australia to New Zealand in two months and another race spreading from western Australia to eastern Australia within a year (Grgurinovic et al 2006). Rust pathogens are in fact intercontinental travellers (Gregory 1963, Viljanen et al 2002 cited in Brown & Hovmoller 2002). Myrtle Rust produces vast numbers of tiny urediniospores which are highly suited to aerial dispersal over long distances. The spore s thick walls resist desiccation and their pigmentation resists ultraviolet radiation allowing them to survive high in the air column Myrtle Rust in the Northern Territory Northern Territory Naturalist (2016) 27 15 for long periods without degradation. Spore longevity is thought to be approximately 90 days (Glen el at. 2007) but would depend on ambient conditions. Thus, vast production of spores and their ability to travel long distances enable the disease to spread rapidly. For example, P. psidii infecting Allspice in |amaica covered an area of 5000 km 2 within one year (Smith 1935 in Glen el al 2007) and in Hawaii the disease spread to all (but one) islands within nine months (Killgore and Hue 2005). In addition to dispersal by wind, Myrde Rust spores are spread by moving infected plant material including nursery stock or cut flowers. At times of movement, plants can appear asymptomatic as the infection may be dormant until conditions arc conducive. Rust spores are also dispersed by human-assisted or animal-assisted means. Spores are inadvertently transported attached to clothing, vehicles, machinery, tools and other equipment (Tommcrup el al 2003) that may come in close proximity to infected plants. Animals such as bees, bats and birds can transport rust spores if they contact infected plant parts during feeding and foraging. Native bees ( Telragonu/a spp.) have been observed harvesting rust spores (possibly due to the resemblance of bright orange spores and pollen) and are thus potentially implicated in transfer of the disease. Detection in Australia Myrtle Rust was first detected in Australia in April 2010 on the central coast of New South Wales in a cut flower nursery (Carnegie el al 2010). Since this initial detection, it had spread to Queensland by late 2010 and to Victoria in 2011 and is now present across much of eastern New South Wales and Queensland. It has also been found in 2015 in northern Tasmania. Detection in the Northern Territory In May 2015, during a routine plant health inspection by Northern Australia Quarantine Strategy (NAQS), officers detected Myrde Rust on Melville Island of the Tiwi Islands, Northern Territory. During the NAQS plant health survey Myrde Rust was observed at four locations over the western part of Melville Island (Fig. 4) on three host species: • cultivated Beach Cherry (Eugenia reinwardtiand) plants; • native mature Utbomyrtus relttsa shrubs (Figs 1,2); and • minor (light) infection on cultivated Weeping Ti-tree (heptospermum madidum) (Fig. 3). Fig. 1. Utbomyrtus retusa shrubs infected by Myrtle Rust ( Pucciniapsidii) on Melville Island, May 2015. (John Westaway) 16 Northern Territory Naturalist (2016) 27 Westaway Fig. 2. Foliage (left) and fruit (right) of Uthomyrtus retusa infected with Myrtle Rust on Melville Island, May 2015. (John Westaway) Of 20 different myrtaceous species inspected on Melville Island in May only these three host species displayed symptoms, with the indigenous l it homy rtus retusa most seriously affected, suggesting this species to be highly susceptible to Myrtle Rust infection. The Northern Territory Department of Primary Industry and Fisheries (DPIF) had been conducting surveillance for Myrtle Rust in Darwin plant nurseries since its arrival in Australia. Following the detection on Melville Island, surveillance was undertaken in nurseries and mainland properties associated with the Tiwi Islands but the rust w'as not found. Highly susceptible plants in Darwin, including Eugenia reinwardtiana and a stand of mature Sysjgium jambos, were checked periodically by the author and found to be symptom free. It was thought the most likely pathway for introduction of the disease was via human agency with nearly all visitors to the Tiwi Islands transiting through Darwin. However an alternative pathway of cyclone-assisted wind dispersal was possible as category 4 Tropical Cyclone Lam passed from Queensland through the Gulf of Carpentaria and onto coastal Northern Territory during the February 2015 wet season, which could potentially have transported Myrtle Rust fungal spores to Melville Island. Fig. 3. Foliage of cultivated Weeping Ti-trce {l^cptospermum madidum) shoving light infection with Myrtle Rust on Melville Island, May 2015. (|ohn Westaway) Such an interstate dispersal event would not be without precedent as Sugarcane Smut (Sporisoriiwi scitamineum) dispersed from Western Australia to Queensland on a particular weather event (Croft el aL 2008) and it is also likely that the fungal Grapevine Leaf Rust that appeared in Darwin in 2001 was a result of wind-born inoculum from Timor-Leste or Indonesia, where the disease is widespread (Daly & Tran-Nguyen 2008). fvlyrtle Rust in the Northern Territory Northern Tjnitory Naturalist (2016) 27 17 puring a plant health survey of Garug Gunak Barlu National Park, Cobourg Peninsula, jn June 2015, NAQS had an opportunity to investigate native and cultivated myrtaceous plant species for symptoms of Myrtle Rust infection. Eighteen different myrtaceous plant species were examined in the field at a range of locations over the eastern parts <>f Cobourg Peninsula and no evidence of Myrtle Rust infection was observed. Plants inspected included the three species found infected at Melville island - Lithomyrtus retusa, Eugenia reinwardtiana and Ijeptospermum madidum — the first two being highly susceptible hosts. Evidence of Myrtle Rust on Cobourg Peninsula would certainly have lent weight to the cyclone pathway hypothesis. In July 2015, DPIF plant biosecurity officers and the author inspected local Darwin populations of Lithomyrtus retusa , the plant severely infected on Melville Island. The nearest populations are located at Berry Springs (lug. 4) and these were found to be infected by Myrtle Rust, albeit more lightly than on Melville Island. Myrtle Rust was subsequently detected on Sysygtum armstrongii in a plant nursery in outer Darwin in September 2015. The infected plants were later destroyed. Syyygutm armstrongii 18 Northern Territory Naturalist (2016) 27 Westaway Fig. 5. Cultivated Eugenia reinwardtiana shrubs infected with Myrtle Rust at the Jingili Water Gardens, September 2015. (John Westaway) had previously been recorded infected by Myrtle Rust (Giblin and Carnegie 2014), but those host plants were presumably cultivated as this species is endemic to the Northern Territory (Northern Territory Herbarium 2015). Populations of 4 . armstrongii occurring in the wild may also be susceptible to infection by P. psidii. Myrtle Rust was also found to have infected two cultivated Beach Cherry (Eugenia reinwardtiana) plants (Fig. 5) at Darwin’s Jingili Water Gardens in late September 2015. Potential Impacts Myrtle Rust infects ‘new growth’, i.e. actively growing shoots and sometimes also buds and fruits (Fig. 2) of susceptible myrtaccous host plants resulting in foliage die- back, reduced photosynthetic and reproductive capacity and increased likelihood of secondary disease. Infection can even lead to tree mortality in some hosts (Carnegie et al. 2015). Potential impacts include economic loss for plant nursery industries growing rust-susceptible varieties of myrraceous plants as Myrtaceae constitute an important component of native plant nursery stock. As mentioned above, Myrtle Rust has been detected on Sytggium armstrongii in a plant nursery situation. There arc some small-scale horticultural enterprises in the Northern Territory that may be vulnerable to Myrtle Rust, for example Guava (Psidiumguajava, P. cattleianum var. cattleianum) crops and edible fruiting trees variously termed Rose/Water/Malay Apple or Jambu (Syeggium aqueutn, S. jambos, S. malaccense, S. samarangense). Currently, Myrtle Rust has not been seen infecting these cultivated Guava or ‘bush apple’ hosts in the Northern Territory. Monocultures of susceptible species are particularly vulnerable. There is potential for nursery Myrtaceae to be treated with appropriate fungicides but this is less feasible for commercial crops such as orchards and silvicultural plantations. For example, commercial bush-food plantations of Aniseed Myrtle (Anetholea anisata) and lxmon Myrde ( Backhousia citriodora) have been affected in New South Wales. Commercial Myrtle Rust in the Northern Territory Northern Territory Naturalist (2016) 27 19 Guava and Eucalyptus plantations are affected in Brazil and the Pimento industry in the Caribbean was devastated by the Myrtle Rust. Amenity plantings arc likely to suffer a decline in aesthetic value as myrtaceous plantings experience dicback. This is potentially significant in Darwin and Palmerston where myrtaceous trees are commonly used in strectside amenity plantings. Further to the direct commercial implications inferred above, there is potential for significant negative impact on economic and spiritual values held by Indigenous people of the Top Find. Two examples are the importance of Sytygium fruit ‘bush apples’ as bush tucker and the spiritual importance of vegetation communities dominated by An-binik (Allosyncarpia ter/iata) in the western Arnhem region (Director of National Parks 2016). Myrtaceae The Myrtaceae is a large and diverse family of trees and shrubs distributed primarily in the southern hemisphere with considerable tropical representation. Although worldwide, the Myrtaceae is particularly significant ecologically in Australia as genera occur in 11 of the 13 major Australian plant formations (Specht 1981) and much of the Australian landscape is characterised by vegetation communities dominated structurally and/or floristically by myrtaceous plants (Pryor & Johnson 1981). On a continental scale, myrtaceous plants represent a high proportion of plant biomass in Australia, and are thus responsible for much of the plant gaseous exchange with the atmosphere and much of the nutrient recycling with soils - both critical environmental services. The Myrtaceae is a key iconic plant family in Australia accounting for approximately 10% of the Australian flora, with more than 2250 species from amongst 95 genera (Australian National Botanic Gardens 2015a). The Myrtaceae contains the greatest number of species of any family of plants in Australia (Beadle 1981, Anonymous 1993) and more than half of the world’s approximately 3000 species of myrtaceous plants are Australian (Australian National Botanic Gardens 2015b). A considerable proportion of this diverse Australian Myrtaceae flora occurs in the relatively moist climatic zone east of the Great Dividing Range that has rainfall and temperature conditions suitable for survival of Myrtle Rust. Puccinia psidii has now been reported from more than 300 Australian Myrtaceae species from 57 genera. Two hundred and thirty species have been infected in the wild and a further 100 by inoculation only (Giblin & Carnegie 2014). The risk Myrtle Rust poses to the conservation of Australian flora is accentuated by the fact that more than Fig. 6. Risk map for the spread of Puccinia psidii. (From Booth and Jovanovic 2012). 20 Northern Territory Naturalist (2016) 27 Westaway 140 species of Myrtaceae are nationally threatened {Environment Protection and Biodiversity Conservation Act 1999; Glen et al. 2007). It is unknown how many of these threatened Myrtaceae are susceptible to Myrtle Rust infection but examples of known susceptible species include Uromjrtus australis and Gossia gonodada, both listed as endangered under the EPBC Act. Twenty-six genera of Myrtaceae are present in the Northern Territory. Many species of these genera occur in the semi-arid and arid zones that are not conducive to persistence of fungal rust infections in general. Pttccinia psidii is unlikely to establish in arid regions due to its requirement for an extended period of leaf wetness (Ruiz et at. 1989 cited in Glen et al 2007). The Top End of the Northern Territory can be defined as the area subject to a tropical monsoon climate (approximating the Territory north of 16 degrees latitude or receiving greater than 600 mm annual rainfall) and the climate here is more similar to the Australian east coast than to arid Northern Territory. I ,ong-tcrm impacts of Myrdc Rust on the natural environment of the Top End arc not known but may be of grave concern as the disease inevitably spreads. A risk map for the spread of P. psidii (Tig. 6) developed by Booth and Jovanovic (2012) depicts the high-risk area of suitable climatic parameters to include the Top End of the Northern Territory. Recent predictive climatic modelling (Krtticos et al. 2013) indicates a low ccoclimatic suitability for P. psidii in a limited area of eastern Arnhem Land and the Tiwi Islands Fig. 7. f .limaie suitability map for Pucciniapsidii in Australia as indicated by the CLIMEX Ecoclimatic Index (Lriticos et at 2013) left, and risk areas identified by Booth et al (2000) for the Northern Territory, right. (Fig. 7), though the authors caution uncertainty concerning the modelled risks in the tropics. These areas coincide with those predicted by the preliminary assessment of Booth et al. (2000) (1% 7). Fortunately the long extended dry season across the Top End of the Northern Territory (with no effective rainfall for 5 to 6 months) may help restrict the spread and impact of Myrtle Rust as it thrives best in humid mesic conditions. The Rust’s ... requirement for leaf wetness ,t UZa u U 2010a) may Umit itS effecavenes s in Strongly seasonal environments a though how this disease behaves in the monsoonal tropics is presently unknown. A p ausible scenario may see Myrtle Rust radiating out from moist sheltered environments unng the wet season and then contracting annually by the harsh conditions of the dry season back to reftiges such as irrigated gardens, spring jungles and riparian vegetation. MMettca or Sjygum species in Top End riparian habitats may however present suitable Myrtle Rust in the Northern Territory Northern Territory Naturalist (2016) 27 21 host and microclimatic conditions for the pathogen to survive the dry season, permitting more or less permanent naturalisation of P. psidii in at least some parts of the Top End. The number of Myrtaceac taxa present in the Top End of die Nordtern Territory can be calculated by subtracting the number of arid zone Myrtaceae (Albrecht et al. 2007) that do not extend their distributions into the Top End from the total Northern Territory Myrtaceae flora (Northern Territory Herbarium 2015; Department of Land Resource Management 2014). This yields some 151 Top End Myrtaceae taxa from 21 genera; with 66 species (of 14 genera) recorded for the Darwin region alone (Dunlop et aL 1995). In contrast to most rusts that infect only a few species. Myrtle Rust is remarkable for its wide host range. Under laboratory conditions about 90% of Australia Myrtaceae tested proved susceptible to Myrtle Rust to some degree (Morin el al. 2011). Given the diversity of species found to be susceptible to Myrtle Rust in Queensland (Giblin & Carnegie 2014, Queensland Government 2015), it seems reasonable to surmise that many/dozens of Top End species arc also likely to be susceptible. Great variation has been observed in the level of susceptibility of myrtaceous plants to this rust ranging from relatively tolerant (e.g. many cucalypts) to extremely susceptible (e.g. Eugenia reimvardtiana , Melaleuca quinquenervia). Some species, e.g. Rbodammia rubescens and Rbodomyrtus psidioides , are impacted to the extent that many individuals die (Carnegie et al. 2015). Most susceptible species however are not killed but their reduced fitness and health arc likely to affect their recruitment capacity'. Susceptibility is also highly variable even among individuals of the same species (Zuaza et al. 2010b; Carnegie et al. 2015). Many widespread common Myrtaceae species of the Top bind (e.g. Eucalyptus tetrodonta, Eucalyptus miniata) as yet show no signs of infection and hopefully this suggests a level of tolerance, supported by the observation that mature cucalypts in eastern Australia, where Myrtle Rust has been established for longer, appear resistant to the disease. However as most Australian Myrtaceae arc naive to rust disease they may yet prove to be susceptible, as pathogens are often more virulent on naive hosts (Glen etaI. 2007). There is possibly a delay time frame before such species become susceptible, perhaps related to Rust strain (biotype), local inoculum loads and mutations rates. Utbomyrtus retusa , the new host record for Myrtle Rust, appears to be especially susceptible based on observations that nearly all individuals (n=100s) inspected on Melville Island were infected (with most being severely infected) (Figs 1—3) whilst all other Myrtaceae (e.g. Eucalyptus , Corymbia, Melaleuca, Lopbostemon, Calytrix) in close proximity m showed no symptoms. The genus ^)ruam, Utbomyrtus has its evolutionary centre in the Northern Territory u — with all but two of the ten species Fig- 8 - Utbomyrtus retusa collections at Australian herbaria. (Map courtesy of AVH) 22 Northern Territory Naturalist (2016) 27 Westaway occurring here, Utbomyrtus retusa is a widespread species across northern Australia (Fig. 8). By contrast, seven of the other eight Utbomyrtus species that occur in the Northern Territory are endemic to the Northern Territory including the fire sensitive Utbomyrtus linariifolia which occurs amongst sandstone outcrops on the western Arnhem Land Plateau. Applying IUCN conservation criteria L. linariifolia is listed under the Territory Parks and If ildlife Conservation Act 2000 as ‘Vulnerable’ to inappropriate fire regimes on account of its obligate seeding regeneration method and also vulnerable to stochastic events due to its small population size (estimated at <1000 mature individuals). It is not known whether U linariifolia is susceptible to Myrtle Rust. The related U obtusa that occurs in coastal Queensland is reported as being susceptible (Giblin & Carnegie 2014) but conspecific status does not appear to confer susceptibility as the diverse list of susceptible versus tolerant Myrtaceac indicates (see Queensland host list, Queensland Government 2015). Although L. linariifolia is the only threatened Myrtaceac species in the Top End there are several restricted range Myrtaceae of conservation value that may be susceptible to Myrtle Rust, including the iconic Arnhem Land monsoon forest dominant tree AHosyncarpia ternata. Allosyncarpia is taxonomically significant as a monospecific genus and .- 1 . ternata is a keystone monsoon forest plant endemic to the specialised geology of the sandstone plateau in western Arnhem Land. Although A. ternata is locally common in sheltered or less fire prone sites on the plateau, its distribution globally is a very limited area. Myrtle Rust has not been found on A. ternata in the wild but the species has been infected in a deliberate inoculation test by CS1RO (Giblin & Carnegie 2014). If Allosyncarpia trees were susceptible, potential impacts may include reduced recruitment and vigour, canopy loss and marginal attrition of the forest community which could expose this Arnhem Land monsoon ecosystem to ftirther fire and weed incursion. Calytrix is another Myrtaceae genus with many endemic or restricted range species in the op End. There are six Calytrix species (C decussata, C. faucicola, C. inopinata , C. min,ana, . rupestris and C. surdiviperana) endemic to the Arnhem Plateau Sandstone Shrubland simplex ecological community (Department of the Environment 2015a), with all but t c rst two being listed with a conservation status of ‘Near Threatened’. It is unknown w ether any of these endemic plants are susceptible to Myrtle Rust. A single Calytrix spectes, C. tetragona from eastern Australia, has tested positive to P. psidii but only by .nocdaOM test, in the wild. The reduced leaf surface area and sderophyllous nature t.a/ytnx may confer some anatomical resistance to infection. f’tfC T< T, End M l racm - b,K, sp. Keep River. , “ f Odrosp.™ Stcnostegpa AsLmyr,,,, tto i S **r*+T *" d * (the first sis listed big Northern Territory endcnucs, are all cons.dered of conservadon concern and listed as “Near Myrtle Rust in the Northern Territory Northern Ttrritory Naturalist (2016) 27 23 Threatened” in the Northern Territory (Northern Territory Herbarium 2015). It is not known whether or not these species are susceptible to infection by Myrde Rust. Rock Myrtle ( Petraeomyrtus putticea) is another key endemic Myrtaceae species of the threatened Arnhem Plateau Sandstone Shrubland Complex ecological community (Department of the Environment 2015b) and its susceptibility to Myrde Rust is also unknown. Vast tracks of the Top End landscape support vegetation comprised of paperbark trees of the Myrtaceae genus Melaleuca , sometimes occurring as monospecific and/or dense stands. There are seven Melaleuca species in the Top End that form extensive vegetation communities, typically on poorly drained or seasonally inundated soils with Melaleuca leucadendra and M. cajuputi amongst the tallest and best formed tree species in the Northern Territory (Dunlop el al 1995) Melaleuca viridiflora , M. cajuputi and Al leucadendra are all of high ecological significance in the Northern Territory as significant character species of several swamp forest, wetland and riparian vegetation communities across the Top End. They could be regarded as keystone species for these communities due to their provision of nectar, pollen, foraging and sheltering substrates and other resources for wildlife such as birds and including migratory species. On account of their community 1 dominance across broad geographic ranges, these Melaleuca species also contribute substantially to the ecological services of water regulation and carbon sequestration. As M. leucadendra is particularly dependent on perennial water sources its potential demise due to Myrde Rust may have negative hydrological and biodiversity repercussions in sensitive riparian habitats. Syeygium armstrongii is another important Myrtaceae tree of Top End riparian habitats and this Northern Territory endemic species has recendy been observed infected with Myrde Rust in a nursery situation. Myrde Rust has been recorded in New South Wales and Queensland on M. viridiflora and M. leucadendra, both of which are rated as ‘highly susceptible’, and also on the closely related Broad-leaved Paperbark ( Melaleuca quinquenervia) which is rated as ‘extremely susceptible’ (Queensland Government 2015). Myrde Rust severely damaged naturalised (introduced) M. quinquenervia in Florida in 1977 (Carnegie & Iidbetter 2012) and has been reported to impact on growth rate and tree structure in eastern Australia (Makinson 2014). Melaleuca viridiflora is an integral component of diverse tropical lowland environments across northern Australia. If indigenous populations of M. viridiflora were to succumb to the effects of this pathogen then there would likely be significant detrimental ecological flow-on effects depending on the degree to which this species is impacted. Even if individual plants are not killed, reduced plant health fitness means less nectar production. Furthermore, their reproductive capacity is likely to be impaired, resulting in lower recruitment and perhaps a slow demise of this significant vegetation community with unknown but likely deleterious implications for its dependant wildlife. 24 Northern Territory Naturalist (2016) 27 Westaway Eucalypts ( Eucalyptus; Corymbia and Angophord) constitute the structural and/or floristic dominant tree species of much of non-arid Australia. Nearly 80 eucalypts (approx. 10% of total) are known to be susceptible to Myrtle Rust though most of these records are from laboratory inoculation tests rather than field observations (Giblin and Carnegie 2014). Furthermore, most mature eucalypts show some resistance or have only a low level of susceptibility. It appears that the vital life stages of seedlings and saplings, as well as cpicormic and coppice growth, are most susceptible to Myrtle Rust infection. This is significant ecologically in Australia for post-fire regeneration and cohort-recruiting species in native ecosystems. Some of the susceptible eucalypts include important forestry species with the major impact for native forestry likely to be on succession, as regenerating seedlings are most vulnerable (Makinson 2014). Across the Top F.nd and perhaps indeed northern Australia, Darwin Stringybark {Eucalyptus tetrodonta) and Darwin Woolybutt ( Eucalyptus miniate, r) arc probably the most prevalent and widespread tree species. It is not understood if either of these two species are susceptible to infection by Myrtle Rust and if so to what degree infected plants may be impacted. Testing in eucalypts indicates there is substantial variation in susceptibility within the same species and between plants from different areas (Zuaza et ai 2010b). Ecological interactions There is likely to be interaction at the plant community level between the pathogen, the plant host and abiotic factors such as climate and fire. The impacts of Myrtle Rust may be most significant in situations where host plants are already stressed due to climatic conditions such as drought, fire regimes, competition from weeds and other factors that have reduced the resilience of the native vegetation communities. Myrtle Rust’s greatest impact may be on plant community succession. If Myrtle Rust hampers regeneration of key or dominant Myrtaceae species thus impeding their ability to compete, there is potential for major changes in plant community- composition at the landscape scale. Such changes would spell habitat loss for native flora and wildlife amounting to fundamental alteration of Australia’s ecology. Poor recruitment and succession resulting in canopy decline may also increase fire impacts and promote invasion of weeds into light or canopy gaps. Furthermore, abiotic consequences such as soil erosion and reduced water retention and quality may be exacerbated. Depending on Rust strain, degree of virulence, environmental conditions and development of tolerance, this disease has the potential to alter the composition and function of forest, woodland, heath and wetland ecosystems. The extended severe dry season conditions typical across the Top End are however not conducive to the prospering of fungal rust pathogens. Top End temperatures may not always suit Myrtle Rust as spore longevity is apparently diminished at temperatures greater than 30°C (Glen et al 2007) and spore germination rates reduce in overnight temperatures greater than 20°C (Kriticos et al. 2013). Myrtle Rust in the Northern Territory Northern Territory Naturalist (2016) 27 25 Susceptibility and impact on host plants may vary into the future as climatic parameters such as rainfall seasonality' change, possibly making some areas more favourable to Myrtle Rust and others less so. The potential for greater impacts may arise if the genetic diversity of the pathogen increases through recombination with novel strains. A genetic/evolutionary ‘arms race’ may ensue between plant hosts developing tolerance and the fungal pathogen evolving to more virulent strains. Due to the relatively rapid reproductive cycle of fungi compared to that of long-lived perennial vascular plants, the odds favour the pathogen. Myrtle Rust cannot be eradicated and will continue to spread, as the fungus produces incalculable numbers of spores that disperse readily via wind, animals and human activity. As there is no practical way to manage the airborne spread of spores, land managers may have to adapt their management practices where possible, for example by addressing other/concomitant pressures such as fire and weeds to alleviate overall impacts. 1 -and managers may be able to utilise management tools such as fire to assist with protection of vulnerable high conservation value vegetation. Though we cannot eliminate Myrtle Rust from northern Australia, we can slow down its spread, manage its impacts and undertake research to discover its full host range whilst seeking longer term solutions. Maintenance and strengthening of quarantine and biosecurity practices to avoid new genetic strains of Myrtle Rust arriving in Australia will help in limiting the pathogens impacts. The Northern Territory Department of Primary Industry and Fisheries has a website with information about Myrtle Rust, and it makes the following recommendations regarding what the public can do to help reduce the spread of Myrtle Rust in the Northern Territory: • avoid importing Myrtaceae plants from New South Wales and Queensland; • if bringing plants in from New South Wales and Queensland, make sure they have been treated with an approved fungicide; and • practise good hygiene when working with plants. Cleaning equipment such as secateurs after use will help reduce the spread of other plant diseases as well. Territory residents and nursery growers are asked to report suspected infected plants by contacting the Exotic Plant Pest Hotline on 1800 084 881. It is also important to avoid plant movements into uninfected areas. If people suspect they have come into contact with Myrtle Rust then careful decontamination of clothing and equipment is required. Makinson (2012) provides detailed advice on appropriate responses to the threat of Myrtle Rust spread including vulnerable asset identification, risk assessment, precautions, decontamination methods, hygiene protocols and options for risk reduction. Acknowledgements 1 am grateful to my colleague Rebecca James, who first noticed Myrtle Rust whilst we were on survey together at Melville Island and for her advice and patience in explaining the strange world of plant disease to a non-pathologist. 1 thank Bob Makinson for 26 Northern Territory Naturalist (2016) 27 Westaway generously sharing his substantial knowledge of Myrtle Rust and its potential impacts. This paper benefited from a thoughtful review by Northern Territory ecologist David Liddle. Thanks also to Northern Territory Department of Primary Industries staff for sharing their data and to Bob Makinson and Richard Willan for encouraging me to prepare this information for the Territory audience. References Albrecht D.E., Duguid A.W., Coulson H., Harris M.G. and Latx P.K. (2007) I 'oscular Plant Checklist for the Southern Bioregions of the Northern Territory: Nomenclature, Distribution and Conservation Status , Second Edition. Northern Territory Herbarium, Alice Springs. Northern Territory Government Department of Natural Resources, Environment and the Arts. Anonymous (1993) Census of Australian Vascular Plants (CAW) Computer Database ()unc 1993). IBIS data network, Australian National Botanic Gardens, Canberra. Australian National Botanic Gardens (2015a) Austra/iun flora ■ statistics, Australian Government, Canberra. < https:/Avww.anbg.gov.au/ausr-veg/ APC-genera-per-family-2010.html> (accessed 15 November 2015). Australian National Botanic Gardens (2015b) Australian flora-statistics, Australian Government, Canberra. (accessed 15 November 2015). Beadle N.C.W. (1981) Origins of die Australian angiosperm flora. In: / Ecological biogeography of Australia (ed. Keast A.), pp. 407-426. Dr W Junk, The Hague. Booth T.H., Old K.M. and Jovanovic T. (2000) A preliminary assessment of high risk areas for Puccinia psidii (Eucalyptus Rust) in the Neotropics and Australia. Agriculture Ecosystems and Environment 82, 295-301. Booth T.H. and Jovanovic T. (2012) Assessing vulnerable areas for Puccinia psidii (eucalyptus rust) in Australia. Australasian Plant Pathology 41, 425—429. Brown J.K.M. and Hovmollcr M.8. (2002) Aerial dispersal of pathogens on the global and continental scales and its impact on plant disease. Science Washington. 297, 537-541. Carnegie A.J., Lidbctter J.R., Walker J., Horwood M.A., Tesoricro I.., Glen M. and Priest M.J. (2010) ( ’redo rangelii, a taxon in the guava rust complex, newly recorded on Myrtaceac in Australia. Australasian Plant Pathology 39, 463-466. Carnegie A. and Cooper K. (2011) Emergency response to the incursion of an exotic myrtaceous rust in Australia. Australasian Plant Pathology 40(4), 346-359. Carnegie A.)., Kathuria A., Pegg G.S., Entwisde P., Nagel M. and Giblin F.R. (2015) Impact of the invasive rust Puccinia psidii (mvrtle rust) on native Myrtaceac in natural ecosystems in Australia. Biological Invasions DOT 10.1007/s10530-015-0996-y. Carnegie A.J. and Lidbctter J.R. (2012) Rapidly expanding host range for Puccinia psidii sensu lato in Australia. Australasian Plant Pathology 41(1), 1.3—29. Croft B.J., Magarcv R.C., Allsopp P.G., Cox M C., Willcox 'LG., Milford B.J. and Wallis E.S. (2008) Sugarcane smut in Queensland: Arrival and emergency response. Australasian Plant Pathology 37, 26-34. Daly A. and Tran-Nguyen L. (2008) Grapevine Leaf Rust - Incursion Risk Analysis and Improvement of PCR Diagnostics (Project ID). Internal Report, Northern Territory Government. Department of the Environment (2015a) Arnhem Plateau Sandstone Shrubland Complex Appendix A. (accessed 9 October 2015). Myrtle Rust in the Northern Territory Northern Territory Naturalist (2016) 27 27 Department of the Environment (2015b) Arnhem Plateau Sandstone Shrubland Complex, Advice to the Minister for Sustainability, Environment, Water, Population and Communities from the Threatened Species Scientific Committee on an Amendment to the List of Threatened Ecological Communities under the EPBC Act. Department of T.and Resource Management (2014) Checklist of the Vascular Plants of the Northern Territory. Northern Territory Herbarium, Department of Land Resource Management. (accessed 16 November 2015). Director of National Parks (2016) Kakadu National Park Management Plan 2016-2026 A living Cultural landscape. Kakadu National Park Board of Management and Director of National Parks Australian Government. Dunlop C.R., 1 .each G.J. and Cowie l.D. (1995) Flora of the Darwin region. Conservation Commission of the Northern Territory, Darwin. Environment Protection and Biodiversity Conservation Act (1999) EPBC Act List of Threatened Flora, Department of Environment and Energy, Australian Government. Ferreira F.A. (1983) Ferrugem do eucalipto. [Eucalyptus rust|. Revista Arvore 7, 91-109. Giblin F. and Carnegie A.J. (2014) Puccinia psidii (Myrtle Rust) — Australian host list. Version current at 24 September 2014. Glen M., Alfenas A.C., Zauza E.A.V., Wingfield M.J. and Mohammed C. (2007) Puccinia psidii: a threat to the Australian environment and economy — a review. Australasian Plant Pathology 36, 1-16. Gregory, P.H. (1963) The spread of plant pathogens in air currents. Advancement of Science 19, 481-488. Grgurinovic C.A., Walsh D. and Macbeth F. (2006) Eucalyptus rust caused by Puccinia psidii and the threat it poses to Australia. EPPO Bulletin 36, 486—489. Kawanishi T., Uematsu S., Kakishima M., Kagiwada S., Mamamoto H., Moric H. and Namba S. (2009) First report of rust disease on Ohia and the causal fungus, Puccinia psidii , in Japan, journal of General Plant Pathology IS, 428-431. Killgore E.M. and Heu R.A. (2005) Ohia Rust Puccinia psidii Winter. New Pest Advisory no 05-04. State of I lawaii. Department of Agriculture: Honolulu. Kriticos D.J., Morin 1,., l-eriche A., Anderson R.C. and Caley P. (2013) Combining a climatic niche model of an invasive fungus with its host species distributions to identify risks to natural assets: Puccinia psidii Sensu lotto in Australia. PlooS ONE 8(5), e64479. doi:10.1371 /journal. pone.0064479 Machado P.D.S., Alfenas A.C., Alfenas R.F., Mohammed C.J. and Glen M. (2015) Microsatellite analysis indicates that Puccinia psidii in Australia is mutating but not recombining. Australasian Plant Pathology 44, 455—462. McTaggart A.R., Roux J., Granados G.M., Gafur A., 'l'arrigan M., Santhakumar P. and Wingfield M.[. (2015) Rust {Pucciniapsidii) recorded in Indonesia poses a threat to forests and forestry in South-East Asia. Australasian Plant Pathology doi:10.1007/sl3313-015-0386-z Makinson R.O. (2012) Myrtle Rust - a new threat to Australia’s biodiversity. A course on Myrtle Rust recognition, reporting, risk assessment, impacts, and management concepts and techniques. Version 3.1. Australian Network for Plant Conservation Inc., in association with the Royal Botanic Gardens & Domain Trust, Sydney. 28 Northern Territory Naturalist (2016) 27 Westaway Makinson R.O. (2014) Key Threatening Process Nomination for: ‘Introduction, establishment, and spread of, and infection by, exotic rust fungi of the order Pucciniales pathogenic on plants of the family Myrraccae’. Confidential unpublished nomination under Environment Protection and Biodiversity Conservation Act 1999 (EPBC Act). Marlatr R.B. and Kimbrough J.W (1979) Puccinia psidii on Pimento dioica in south Florida. Plant Disease Reporter 63, 510-512. Morin L., Aveyard R. and Lidbcttcr |. (2011) Myrtle rust: host testing under controlled conditions. CSIRO Ecosystem Services and NSW Department of Primary Industries. Northern Territory Herbarium (2015) I'/oraNTNorthern Territory P/ora Online. Department of Land Resource Management, http://eflora.nt.gov.au (accessed 16 November 2015) Pryor L.D. and Johnson L.A.S. (1981) Eucalyptus, the universal Australian. In: Ecological biogeography of Australia (ed. Keast A.), pp. 499-536. Dr W Junk, The Hague. Queensland Government (2015) Known plants affected by Myrtle Rust. Ramsficld T., Dick M., Bulman L. and Ganlcy R (2010) Briefing document on Myrtle Rust, a member of the Guava Rust complex, and the risk to New Zealand. SCION Research (Scion New Zealand Crown Research Institute, Rotorua NZ). Roux J., Grcyling I., Coutinho T.A., Verlcur M. and Wingfield M.J. (2013) The Myrtle rust pathogen, Puccinia psidii, discovered in Africa. IMA Fungus 4(1), 155—159. doi:10.5598/ imafungus.2013.04.01.14 Ruiz R.A.R., Alfenas A.C., Ferreira F.A. and do Vale F.X.R. (1989) Influence of temperature, time of leaf wetness, photoperiod and intensity of light on infection by Puccinia psidii in Eucalyptus. Fitopatologia Brasi/eira 14, 55-64. Simpson J.A., Thomas K. and Grgurinovic C.A. (2006) Urcdinalcs species pathogenic on species of Myrtaccac. Australasian Plant Pathology 35, 549-562. Smith F.E.V. (1935) Rust Disease of Pimento. The journal of the Jamaican Agricultural Society 39, 408-411. Specht. R.L. (1981) Major vegetation formations in Australia. In: Ecological hiogeograptry of Australia (ed. Keast A.), pp. 163—297. Dr W Junk, The Hague. I'ommerup I.C., Alfenas A.C. and Old, K.M. (2003) Guava rust in Brazil - a threat to Eucalyptus and other Myrtaceae. New Zealand journal of Forestry Science 33, 420-428. Uchida J., Zhong S. and Killgore E. (2006) First Report of a Rust Disease on Ohia Caused by Puccinia psidii in Hawaii. Plant Disease 90, 524. \ iljanen Rollinson S.L.H. and Cromey M.G. (2002) Pathways of entry and spread of rust pathogens: Implications for New Zealand’s biosecurity. New Zealand Plant Protection 55, 42—48. Winter G. (1884) Reper/orium. Rabenhors/ii fungi europaei et ex/raeuropaet. Cent. XXXI et XXXII. Hedwigia 23,164-172. Zauza E.A.V., Couto M.M.F., Lana V.M. and Maffia L.A. (2010a) Vertical spread of Puccinia psidii uredimosporcs and development of eucalyptus rust at different heights. Australasian Plant Pathology 39, 141—145. Zauza E.A.V., Couto M.M.F., Lana V.M. and Maffia L.A. (2010b) Myrtaccac species resistance to rust caused by Puccinia psidii. Australasian Plant Pathology 39, 406—411. Zhuang J. A and Wei S.-X (2011) Additional materials for the rust flora of Hainan Province, China. Mycosystema 30(6), 853-860. Northern Territory Naturalist (2016) 27: 29—35 Short Note Asystasia gangetica subsp. micrantha, a new record of an exotic plant in the Northern Territory John O. Westaway 1 , Lesley Alford 2 , Greg Chandler 1 and Michael Schmid 2 1 Northern Australia Quarantine Strategy, Commonwealth Department of Agriculture, 1 Pederson Road, Marrara, NT 0812, Australia Email: iohn.westaway@agriculture.eov.au 2 Veg North, PO Box 124, Nightcliff, NT 0814, Australia Abstract An herbaceous weed of the acanthus family, Asystasia gangetica subspecies micrantha, sometimes known as Chinese Violet, was found naturalised in Darwin in April 2015 and was immediately eradicated. Although cultivated as an ornamental, this plant is regarded as an invasive weed in eastern Australia where it has been established for 15 years, and is a recognised problem weed in neighbouring tropical countries. Identification and taxonomic aspects of this species are briefly discussed, as is its distribution in Australia and overseas, and its possible means of arrival in Darwin. Introduction Asystasia gangetica subspecies micrantha (Nees) Ensermu is a target weed species of the Northern Australia Quarantine Strategy which means that it has been identified as a plant that, if introduced, is likely to have substantial detrimental impacts on agricultural production and the environment. Asystasia gangetica subsp. micrantha is also on the /\lert List for Environmental Weeds (Australian Government Department of Environment 2000), a list of non-native plants that threaten biodiversity and cause other environmental damage. Asystasia gangetica subsp. micrantha is a form of Chinese Violet and belongs to the large, predominantly tropical plant family Acanthaccac. It is a perennial herb that can grow in a mat-forming habit and smother more desirable ground plants, thus potentially affecting agriculture or reducing biodiversity. It is a major weed overseas, particularly in Malaysia, Indonesia and the Pacific Islands (Kiew & Vollesen 1997; Anonymous 2003; Hsu et al. 2005). In these places it infests plantations and competes effectively for soil nutrients, especially nitrogen and phosphorous (Barnes & Chan 1990: 148), reducing productivity and increasing crop management costs. It could also become an agricultural weed in Australia. The taxonomy of Asystasia requires worldwide revision and A. gangetica is a variable species with two subspecies recognised, vi% subsp. gangetica and subsp. micrantha. 30 Northern Territory Naturalist (2016) 27 Westaway et al. Fig. 1. Habit of Arystasia gangetica subsp. micrantha at Rapid Creek, JOW 4806. (John Westaway) Fig. 2. Flowers of Asystasia gangetica subsp. micrantha illustrating the distinctive purple markings on the lower corolla lobe. (John Westaway) Detailed botanical descriptions of A. gangetica subsp. micrantha arc available in Rnsermu (1994) and Kiew & Vollcscn (1997), but in summary this plant is a sprawling perennial herb that grows rapidly to 0.5 m tall (higher on supporting vegetation) and can form mats due to its propensity to take root at stem nodes (Fig. 1). I .eaves are oval shaped (to approx. 15x5 cm), opposite, and paler on the lower surface. The bell-shaped flowers (usually 15-25 mm long) are white with distinctive purple blotches in two parallel lines (Fig. 2). Club-shaped seed capsules (approx. 30 mm long) have four flattened seeds attached by hooks. The two subspecies are closely related but differ in floral morphology with the typical subspecies having larger flowers (greater than 30 mm long) that may be blue, mauve, white or sometimes yellow but lack the purple blotches on the lower corolla lobe. Asystasia gangetica subsp. micrantha in the NT Northern Territory Naturalist (2016) 27 31 Asyslasia gangetica subsp. gangetica is sometimes cultivated in tropical gardens for its large showy flowers and has also become naturalised in the Northern Territory and Queensland. The two subspecies also differ in their ecology with subsp .gangetica typically found as a relatively benign ‘garden escapee’ which does not seem as successful or aggressive as subsp. micrantha at invading bushland. Distribution Asystasia gangetica subsp. gangetica occurs from India to SE Asia but is cultivated more widely in tropical zones. This typical subspecies sometimes forms naturalised populations beyond gardens and has been recorded in settlements across the Northern Territory from Darwin to Crokcr Island, Maningrida, Numbulwar and Dhalinybuy in eastern Arnhem Land. Asystasia gangetica subsp. micrantha is native only on the African continent, but is also cultivated and is now widely naturalised in Asia, the Pacific and central and southern America (for example, sec Kiew & Vollesen 1997; Gorham & Hosking 2003; PIER 2006; Daniel & Eigueiredo 2009; Lujan et al. 2012). It can be found as naturalised populations in neighbouring countries of biosecurity interest to Australia such as Indonesia, Timor-Leste, Papua New Guinea and the Solomon Islands. There are three main areas of establishment of A. gangetica subsp. micrantha in Australia (Fig. 3). It was first encountered in Australia in the Port Stephens area in 1999 (on the New South Wales mid-north coast) where it is established at a number of locations in and near Anna Bay and Boat Harbour (Anonymous 2003, Skinner 2015). The species was later found at Shoal Water Bay Training area, north of Rockhampton in Queensland in 2011 where there are at least tw'o naturalised populations established on this military land (HERBRECS). It is also established on the Gold Coast in south east Queensland by 2013 where it has ingressed into native vegetation in Currumbin Conservation Park (HERBRECS; Anonymous 2014). Asystasia gangetica subsp. micrantha was found naturalised near the Darwin airport in April 2015 by Michael Schmid and Lesley Alford of Veg North whilst they were conducting weed management work in Darwin International Airport’s Rapid Creek Reserve (Fig. 4). This represents the first confirmed naturalised population of this taxon in the Northern Territory. It was present on Darwin International Airport land adjacent to the drain Fig. 3. Distribution of exotic occurrences of Asystasia gangetica subsp. micrantha in Australasia as indicated by herbarium records stored in the Australian Virtual I lerbarium. New Northern Territory record in green. 32 Northern Territory Naturalist (2016) 27 Westaway et a!. Fig. 4. The habit of Asystasiagangetica subsp. micrantha at Rapid Creek (left), and following weed treatment (right). (Lesley Alford, John Westaway) running north to Rapid Creek from behind the Rydges Darwin Airport Resort barbecue area (12.4041 °S, 130°.8807’E). Identification was confirmed by [W and a specimen has been lodged at the NT Herbarium (JOW 4806). At this site many sprawling plants together occupied approx 5—10% ground cover (Fig. 4) over an area of about 60 nr in mulched brown clay loam at the edge of a minor drainage channel (Fig. 4). The vegetation there consisted of Acacia auriculiformis and Corymbia be/la — dominated remnant grassy woodland with Pandams spiralis common in the understory. Asystasia gangetica subsp. micrantha had apparendy been cultivated in the George Brown Darwin Botanic Gardens and subsequently eradicated (Anonymous 2003). There is an horticultural record of Asystasia gangetica from 1994 at these Gardens but the subspecies is not specified (Ben Wirf pers. comm.). Should A. gangetica subsp. micrantha became more widely established in Australia, it could potentially impact commercial agricultural crops such as vegetables, legumes, cur flowers and horticultural and forestry (e.g. Fig. 5) enterprises (Anonymous 2003; Skinner 2015). Skinner (2015) provides a summary of this species impact and its management in Australia and internationally. Its competitive success over a wide geographical range is attributable to its fast establishment, stoloniferous growth form capable of rooting at stem nodes, rapid growth rate, early flowering and high seed production (Anonymous 2003). As an environmental weed, it is likely to have similarly detrimental effects, smothering native Fig. 5. A. gangetica subsp. micrantha completely dominating the field layer in a teak forest in western java. (John Westaway) Asystasia gangetica subsp. micrantba in the NT Northern Territory Naturalist (2016) 27 33 flora and degrading wildlife habitat, particularly in already modified environments. The species’ ability' to readily invade undisturbed native vegetation is unclear. Asystasia plants at Rapid Creek were treated with the herbicide 2,4-D as part of proactive weed work being undertaken by Darwin International Airport who have a process in place for management of the environment of their leased lands. Taxonomy In a paper on morphological variation within Asystasia gangetica , Ensermu (1994) recognised two subspecies; one the large-flowered type of the species from India, throughout Asia to Indonesia and Pacific Islands, and the other a smaller-flowered African taxon, subsp. micrantba. Kiew & Vollesen (1997) tabulated morphological differences between the two subspecies. However the NSW Herbarium (PlantNct) and the Australian Plant Census (APC), an authoritative conspectus of Australian plant taxonomy, have recently relegated the two subspecies to synonymy under A. gangetica. The Northern Australia Quarantine Strategy (NAQS) has observed and collected both subspecies in several locations in northern Australia and neighbouring countries during the course of plant health surveys and is of the view that the two subspecies are distinct on the basis of differing floral morphology and different invasion behaviour. NAQS has been conducting a molecular study to investigate genetic variation between the two subspecies (and closely related congeners) across the geographic range of A. gangetica subsp. micrantba. Preliminary results from this analysis, which will be published m the near future, strongly support the contention that A. gangetica subsp. micrantba is distinct from the typical subspecies. This is further supported by the existence of a difference in the number of chromosomes in the two taxa, with subsp. gangetica being tetraploid (four sets of chromosomes) and subsp. micrantba diploid (two sets of chromosomes). This creates reproductive isolation between the two taxa, with any hybridisation resulting in the production of sterile triploid plants (three sets of chromosomes). Introduction to Australia Asystasia gangetica (especially subsp. gangetica) is cultivated in tropical climates as an ornamental garden plant. It spreads by seeds released explosively from drying capsules, as well as by stems which are capable of taking root when in contact with moist soil. Dispersal over long distances is by human agency through accidental transportation of plant material in gardening, landscaping, roadworks, mining and defence activities. The original incursion of A. gangetica subsp. micrantba in the Port Stephens area is thought to be derived from an horticultural introduction with subsequent populations having also spread from garden plantings or resulting from the dumping of garden waste. Care is required to ensure correct disposal of plant material as much of its spread has been attributed to poor disposal of plant parts which can contain seeds and broken stems that can readily establish and form new plants, causing the infestation to spread. 34 Northern Territory Naturalist (2016) 27 Wes taway et al. The means of introduction of the recent Darwin incursion at the airport may be via seed inadvertently transported by people or machinery from eastern Australia or overseas. A plausible scenario could entail a visitor from overseas staying at the cottages along the back of the resort and accidently depositing seed from his/her footwear or clothing that had travelled with them. The infestation occurred in a drain that runs off the airport so it is possible that plant material could have arrived on an aircraft, or machinery transported by air. It is perhaps no coincidence that the Darwin incursion is near Defence land as is the Shoalwater Bay site in central Queensland, suggesting that transport via military hardware may have been implicated. The Darwin population at Rapid Creek has been treated with herbicide and any regeneration will be monitored and treated as necessary. Soil removed from the site during drain maintenance is disposed of appropriately to ensure seed and plant material is not spread. Whether propagulcs were transported beyond the site prior to treatment (e.g. downstream in die Rapid Creek catchment) remains to be seen next wet season. Acknowledgments LA & MS thank Chris Collins of the Northern Territory Weed Management Branch for discussion of management of this species. References Anonymous (2003) Chinese violet {Asystasia gangetica ssp. micranthd). Weed Management Guide. CRC for Weed Management, Australia, 6 pp. (accessed 7 August 2015). Anonymous (2014) Weed Spotters’ Network Queensland Bulletin, February 2014. Department of Science, Information Technology, Innovation and the Arts. 8 pp. (accessed 24 September 2014). Australian Government Department of Environment (2000) Alert List for Environmental Weeds website, (accessed 7 August. 2015). AVH [Australian Virtual Herbarium], online plant distribution database. Adas of Living Australia. (accessed 7 August 2015). Barnes D.E. and Chan L.G. (1990) Common weeds of Malaysia and their control. Ancom, Shah Alam, Malaysia. Daniel T.E and Figueiredo E. (2009) The California Academy of Sciences Gulf of Guinea Expeditions (2001,2006,2008) VII. Acanthaceae of Sao Tome and Principe. Proceedings of the California Academy oj Sciences, Series 4 60, 623—674. Ensermu k. (1994) A revision of Asystasiagangetica (L.) T.Anders. (Acanthaceae). In: Proceedings of the / 3th Plenary Meeting, Aetfat, Zomba, Malawi, Plants for the People VoL 1 (eds SeyaniJ.H. and Chikuni A.C.), pp. 333-346. Asystasia gangetica subsp. micrantha in the NT Northern Territoiy Naturalist (2016) 27 35 Gorham P. and HoskingJ. (2003) Weed Alert: Have you seen this plant? A form of Chinese violet Asystasia gangetica subspecies micrantha. (NSW Department of Primary Industries, Sydney) (accessed 23 September 2015). HERBRECS Queensland Herbarium database; accessed via AVH [Australian Virtual Herbarium]. Hsu T.-W., ChiangT.-Y., PengJ.-J. (2005) Asystasia gangetica (1,.) T. Anderson subsp. micrantha (Nees) Ensermu (Acanthaceae), a newly naturalized plant in Taiwan. Taiwania 50,117—122. Kicw R. and Vollesen K. (1997) Asystasia (Acanthaceae) in Malaysia. Kcw Bulletin 52: 965-971. Lujan M., Gutierrez N. and Gaviria J. (2011) Asystasia gangetica (L.) T. Anders, subsp. micrantha (Nees) Ensermu (Acanthaceae), un nuevo reporte para Venezuela. Ernstia 21,131-137. Pacific Island Ecosystems at Risk (PIER) (2006) Asystasia gangetica (L.) T. Anderson, Acanthaceae. Plant threats to Pacific ecosystems. Institute of Pacific Islands Forestry, Hawaii, USA. (accessed 24 September 2015), PlantNet NSW FloraOnline (accessed September 21, 2015) Skinner, J. (2015) The invasive weed Chinese violet [Asystasiagangetica subspecies micrantha) now threatens northern Australia. Plant Protection Quarter^ 30(4), 126-132. Northern Territory Naturalist (2016) 27: 36-46 Research Article Seed viability of native grasses is important when revegetating native wildlife habitat Sean M. Bellairs and Melina J. Caswell Research Institute for the Environment and Livelihoods and School of Environment, Charles Darwin University, Darwin, NT 0909, Australia Email: sean.bellairs@cdu.edu.au Abstract Native grasses are a dynamic and essential component of the majority of terrestrial ecosystems in the Northern Territory. Restoring native grasses in disturbed environments is important for providing faunal habitat, reducing surface erosion and resisting weed invasion. However, establishing native grasses has been problematic in many regions of Australia due to seed viability issues. We investigated 48 seed lots of 29 Northern Territory native grass species to determine whether seed quality was an issue for establishment of tropical native grasses. Seed lots were largely collected by commercial seed suppliers, rather than by research staff, so the samples reflect seed lots that could be sourced for revegetation projects. The seed purity, proportions of filled seeds, visually viable seeds and mctabolically active seeds were assessed. Viability responses to storage were investigated in 15 seed lots. The proportion of florets that contained a seed (caryopsis) ranged from 10-97% (average 62%) and between 0—79% of the florets contained metabolically active seeds (average 36%). Two seed lots had viability' of 0—10% and 12 of the 48 seed lots had less than 30% seeds that were metabolically active and potentially viable. Thus, seed quality limits establishment of tropical native grasses from sown seeds in the Northern Territory. When using native grasses to establish native habitat it is important to assess the quality of the seeds and use a sufficient quantity of seeds for effective establishment of these grasses. Seeds of many species will maintain viability' for several years if stored in cool dry' conditions. Seed for revegetation projects can therefore be collected and stored over several years. Introduction Native grasses are a feature of the vegetation communities of the Northern Territory. Spear grasses (Heteropogon and Sorghum spp.) and spinifex (Triodia spp.) arc particularly dominant grasses in tropical and arid vegetation communities. In tropical communities there is typically a considerable diversity of other native grass species present. Native grasses provide a range of valuable ecological functions including: • providing food for granivorous mammals and birds; • providing habitat for native fauna; • resisting invasion by introduced weed and improved pasture species; and • assisting with control of surface erosion. Seed viability of native grasses Northern Territory Naturalist (2016) 27 37 When re-establishing native vegetation communities during revegetation activities, the establishment of native grasses is often problematic. Active establishment of native grasses as part of revegetation activities generally relies on sown seeds, and those seeds need to be of good quality. In many regions of Australia seed biology issues relating to seed viability and seed dormancy often limit establishment of sown native grasses. We suggest this is also the case in the tropics of the Northern Territory. The seed of a grass is termed a caryopsis and this is typically enclosed within two sheathing covering layers (the palea and lemma) to create a floret and several florets are enclosed within two glumes (Tig. 1). The seed of a native grass may be dispersed cither Fig. 1. Mature ‘seeds’ of Cockatoo Grass (Alloteropsis semiakta ) showing the intact spikelets (middle) and the actual seeds (caryopscs) after extraction from the covering structures (top). The bottom spikelct has been opened to remove the caryopsis and the remains of the outer glumes and the inner lemmas covering the two florets can he seen. Opening or cutting the spikelet to check for a filled caryopsis is an easy initial test for checking the proportion of spikelets that contain viable seeds. as a bare caryopsis or enclosed within the floret. Seed lots of native grasses generally contain florets, other inflorescence material such as the stigma and stamens (Tig. 2), and sometimes vegetative material, as well as the caryopscs (Tig. 3). There may be a low proportion of viable seeds in a seed lot if the seed lot contains a high proportion of vegetative material, a high proportion of empty florets, or a high proportion of damaged or dead caryopses. A simple indicator of poor seed fill is to lightly press on the sides of the florets to feel the caryopsis. Alternatively, the floret Fig. 2. Cockatoo Grass (Alloteropsis semialata) (above) and Giant Spear Grass (/ leteropogon triticeus) (below) flowering. The Cockatoo Grass flowers have orange and yellow stamens and purple feather-like stigmas. The Giant Spear Grass has purple stamens and long brown awns protrude from the apex of the inflorescence. 38 Northern Territory Naturalist (2016) 27 Bellairs & Caswell can be cut in half to visually inspect the caryopsis. Sometimes caryopscs may be of normal size and appearance but still be dead. To test viability in this case, a sample of the caryopses can be treated with tetrazolium chloride, a dye that will turn red if the tissues of the caryopsis are metabolically active (Merritt 2006). Poor seed viability can be due to site factors affecting the grass plant, such as adverse seasonal growing conditions, habitat features that do not suit the species, or damage by fungi or insects. Some species are genetically disposed to poor seed production (Jacobs 1973). When collecting the seeds, seed maturity' and collecting techniques can affect viability'. After collection, storage conditions (including temperature, humidity, fungi and insects) affect the rate of deterioration of seed quality' (Merritt 2006). We investigated 48 seed lots of 29 native grass species to determine whether seed quality' was an issue for the establishment of native grasses in the Top End of the Northern Territory. The proportions of filled seeds, seeds of normal appearance and metabolically active seeds were assessed. Viability responses to seed storage in an air conditioned room were investigated by repeated testing of 15 of the seed lots. Materials and Methods Seeds were mainly supplied by Greening Australia NT and by Kakadu Native Plant Supplies, with a list of desired species sent to their collectors. Seeds were collected in rlw Jabiru, Darwin and Katherine regions. Four seed lots (A lloteropsis senna lata Lot 3; EriackrtP ciliata 1 .ot 3, Eriachne schultvjana Lot 2 and Tbaumastocbloa major I -ot 1) were collected wit h assistance from Charles Darwin University (CDU) staff and/or students. Seeds were; collected between .April 2005 and May 2011. The date or month of collection provide^ by the seed collector or seed supply company and the date of testing were recorded - Fig. 3. Seeds and associated structures of Giant Speargrass ( Heteropogon trilicens) (top left). Kangaroo Grass Qhcmeda Iriatidra) (top right). Love Grass ( Eragrostis spartinoides) (bottom left) and Wanderrie grass ( Eriachne schulfjana) (bottom right). The small brown seeds or caryopscs extracted from the covering structures are shown for Love Grass and Wanderrie Grass. Seed viability of native grasses Northern Territory Naturalist (2016) 27 39 Stored seeds were accepted but we requested that new seed lots were sent to CDU as soon as possible after collection and cleaning of florets from vegetative material. Most seed lots were tested within two months of arrival at CDU. Four of the 48 seed lots were not tested until 9-10 months after arrival and five seed lots were tested 14—16 months after being received ( Aristida inaequiglum'ts Lot 2, Eulalia aurea, Heteropogon contortus, Sorghum plumosum, S. timorense). 1 f sufficient seeds were available, seed lots were resampled and retested after one or two years of storage. Seed purity refers to the weight of undamaged florets and caryopses as a proportion of the total weight of the seed lot. Seed purity was assessed by removing all vegetative material, chaff and obviously damaged florets from undamaged florets and caryopses for small samples. Larger seed lots were sub-sampled using halving techniques prior to assessing purity of four sub-samples and the purity result is the average of those four sub-samples. Seed fill and cut tests were conducted using four replicates of 25 florets. Seed fill data denoted the percentage of florets that contained a caryopsis within them when the florets were opened. For the cut test, the caryopsis w'as removed from the floret and inspected under a dissecting microscope, where the percentage of florets with visually viable caryopses w'ere counted. Unfilled florets, shrivelled, discoloured or damaged caryopses and caryopses that had a missing embryo were assessed as not viable. Possibly viable seeds that were only slightly smaller or slightly discoloured were counted as viable and included in the tetrazolium assessment below. E ragrostis spartinoides did not have seed fill of florets assessed as the seeds disperse as caryopses and don’t retain the outer floret structures. For those seeds that were visually assessed to be viable or possibly viable, 2,3,5 triphenyl tetrazolium chloride (TTZ) was used to determine any metabolic activity. This colourless solution becomes red in response to metabolic activity in the tissue. Four replicates of 25 seeds were preconditioned by placing them in water at room temperature for 24 hours. The covering structures were removed or pierced away from the embryo to ensure water uptake without causing damage to the embryo. After imbibition, the seeds were cut through the embryo (or close to the embryo if cutting caused damage) except for Eragros/is spartinoides seeds, which were too small to cut. Half of each seed was then placed into 1% TTZ solution in a glass vial covered with aluminium foil to keep the incubating seed in darkness. The vials were placed in an incubator at 30°C for 24 hours, after which the seeds were removed and inspected under a dissecting microscope. Seeds with deep red-stained embryos and storage tissues were considered viable. Seeds that were unstained were not viable. Seeds with the embryo stained pale pink, or mottled staining of the storage tissues, were considered possibly viable. This resulted in a minimum and maximum proportion of viable seeds as assessed by TTZ. 40 Northern Territory Naturalist (2016) 27 Bellairs & Caswell Results Seed viability of the grass species was variable, with Aristida inaequiglumis Lot 2 and Themeda triandra Lot 1 having less dian 2-6% of viable florets, whereas Heteropogon triticeus Lot 1 and Eragros/is spartinoides had 79% and 90—100% viable seeds respectively (Table 1). Average viability of all grasses was moderate (37—42%). Table 1. Viability of native grass seed lots. Lach row is a separate seed lot. Age is the approximate time in weeks from collection to testing. Purity’ is the proportion by weight of caryopses plus florets that contain caryopses relative to the seed lot weight. Seed fill is die proportion of florets containing a caryopsis. Cut test is die proportion of florets containing a visually viable caryopsis. The last columns are the minimum and maximum viability of florets after tetrazolium (TTZ) testing as a proportion of florets (or caryopses if caryopses arc shed from the florets). Species Age (weeks] Purity (%) Seed fil (%) Cut test (%) TTZ min (%) TTZ max (%) Cockatoo Grass 5 98 53 35 25 33 AHoteropsis semiatata (R.Br.) Hitchc. 38 94 71 56 26 36 4 61 20 41 Kerosene Grass Aristida bolathera Domin 34 35 97 94 65 76 Feathertop Threeawn 52 66 44 44 24 35 Aristida inaequiglumis Domin 91 60 45 37 0 2 Golden Beard Grass 58 68 85 68 34 46 C/irysopogon fa/tax S.T.Blake 39 88 66 56 50 53 Ribbon Grass Cbrysopopon latifolius S.T. Blake 45 91 52 24 14 18 Silky Heads Cymbopoepn bombycinus (R.Br.) Domin 38 59 70 58 56 56 Queensland Blucgrass 40 86 43 34 30 31 Dichanthium serictum (R.Br.) A.Camus 144 79 43 41 25 29 Hare’s Foot Grass Ectrosia leporina R.Br. 39 55 55 49 49 49 Love Grass Eragpostis spartinoides Steudcl 56 8 n/a 100 90 100 Wanderrie Grass 10 50 59 49 49 49 hriachne qgrostidea F.Mucll. 18 56 65 41 34 37 Longawn Wanderrie Grass 20 32 67 46 37 41 Eriacbne armitii F.Muell. ex Bcnth. 41 51 52 37 33 35 Wanderrie Grass Eriacbne avenacea R.Br. 94 80 64 60 59 59 Wanderrie Grass) 112 100 78 39 23 33 Eriacbne burkittii jansen 11 78 49 20 13 14 Slender Wanderrie Grass 12 32 60 26 16 18 Eriacbne ciiiata R.Br. 9 42 89 53 48 49 Pan Wanderrie Grass Eriacbne gfauca R.Br. 31 100 74 42 19 26 Continued on next page Seed viability of native grasses Northern Territory Naturalist (2016) 27 41 Continued from previous page Species Age weeks) Purity 5 (%) >ced fill ( (%) 2ut test <%> TTZ min (%) TTZ max (%) Northern Wanderrie Grass 16 100 51 41 25 32 Eriachne oblusa R.Br. 39 72 53 36 41 44 15 82 62 49 38 42 Wanderrie Grass 63 60 83 81 15 24 Eriachne scbultzjana F.Muell. 4 - 80 80 24 50 Wanderrie Grass 56 30 42 40 27 34 Eriachne /rise/a Nees ex Steud. 95 17 40 27 26 27 Silky Browntop Eulalia aurea (Bory) Kunth 94 67 23 13 10 12 Black Speargrass Heteropogou contortus (L.) P.Beauv. ex Roem. & Schult. 92 36 73 65 27 36 Giant Speargrass - 100 91 83 79 79 Heteropogou triticeus (R.Br.) Stapf 91 27 89 75 70 72 40 52 75 29 20 26 21 70 62 38 30 36 Fire Grass Schizachyrium fragile (R.Br.) A.Camus 50 68 70 59 53 56 Pigeon Grass Setaria apiculata (Scribn. & Merr.) K.Schum. 126 100 87 73 53 64 Darwin Canegrass Sorghum irttrans F.Muell. ex Benth. 92 55 69 61 59 60 Plume Sorghum Sorghum plumosum (R.Br.) P.Beauv. 92 46 57 56 51 52 Black Soil Canegrass Sorghum timorense (Kunth) Buse 41 68 89 83 62 66 Thaumastoch/oa major S.T.Blake 4 47 65 48 41 47 18 76 80 69 69 69 Kangaroo Grass 37 9 10 6 0 6 Themeda triandra Forssk. 45 24 57 40 40 40 29 13 72 51 46 49 Curly Spinifex Triodia bitextura Lazarides 40 65 21 20 17 18 Mean of all seed lots 47.7 60.7 62.6 49.6 36.7 41.8 Minimum 4 8 10 6 0 2 Maximum 144 100 97 100 90 100 Seed purity was highly variable, ranging from 8-100%, but purity is dependent on the level of cleaning. The Eragrostis sp. seed batch had only 8% pure seeds but the seeds were tiny and numerous - the 17.8g seed lot still contained 24,200 seeds. Seed purity can also be highly variable between seed lots of a species; Ueteropogon triticeus 1 ,ot 1 contained 100% pure seed whereas Lot 2 contained only 27% pure seeds. 42 Northern Territory Naturalist (2016) 27 Bellairs & Caswell Once chaff, vegetative material and damaged florets had been discarded, seed viability was largely dependent on the seed fill of the florets. On average, 37% of florets did not contain seeds. However, the proportion of filled seeds could be much lower — 90% of the florets of The me da triandra Lot 1 did not contain caryopscs and 79% of Triodia bitextura florets were empty. In contrast, other seed lots contained a high proportion of filled seeds - 72% of florets of Themeda triandra Lot 3 and 97% of die florets of Aristida holathera contained caryopses. Closer inspection of the seeds was important for assessing viability as some filled seeds were not viable. Across all the seed lots, 62.6% of florets contained a caryopsis but only 49.6% of the caryopscs were of normal appearance and contained an undamaged embryo. For the two seed lots of Eriachne burkittii, closer visual inspection determined that less than half of the filled florets contained viable caryopses. In contrast for the Eriachne schult^iana florets, almost all of caryopses present appeared viable after microscopic inspection. Some caryopses that appeared normal and viable when visually inspected were not viable as they were not metabolically active when tested with TTZ. For some species, such as Eriachne schult^iana, visual inspection substantially overestimated the number of viable seeds, with 81% of seeds appearing viable in Lot 1 when inspected, but only 24% having any metabolic activity. For other species, such as Cymbopogon bombycinus, all the seeds that visually appeared to be viable were also metabolically active. For the seed lots that were retested after at least one year of storage in the CDU laboratory, the decline in viability averaged 6% per year or a loss of 12% of viable seeds per year (Fig. 4). For some species, such as Eriachne schultvpana, there was substantial reduction in viability after 1-2 years but for others, such as Eragrostis sparlinoides, there was little decline in viability. The Eragrosfis sparlinoides seed lot investigated in this study maintained viability and germination levels after 4.5 years storage and hence some native grass seed lots are able to be stored for quite some time. Eriachne obhtsa and Chrysopogon fallax seed lots were able to be stored for two years with similar levels of viability to the initial assessments, however, after 3-4 years storage viability' levels were low. Eriachne schultspana and Heleropogon triticeus were able to be stored for 1—1.5 years, however, after further storage viability levels were low. Alloteropsis semialata seeds lost viability after 2 years storage with very low viability after 3 years. Discussion Generally, native grass seed lots newly received from seed suppliers in the Darwin region of the Northern Territory had reasonable levels of seed viability, similar to those for Australian native grass seed lots generally. Viability of four native grasses used for mtnesite rehabilitation in Western Australia was generally lower, ranging from 19-39% (Dixon 1997) and three seed lots of Themeda australis from New South Wales ranged from 52—68% (Nolan et al. 1997). Farley et al. (2013) assessed viability of 13 native grass Seed viability of native grasses Northern Territory Naturalist (2016) 27 43 2 . .a re *5 E D E 1 s 100 90 80 70 60 50 40 30 20 10 0 — 5-V *«V (accessed 14Jan 2015). Jacobs S.W.L. (1973) Ecological studies on the genera Triodia R. Br. and Plcctrachnc Henr. in Australia. Unpublished PhD thesis. University of Sydney, Sydney. Loch D., Adkins S., Hestlchurst M., Paterson M. and Bellairs S. (2004) Seed formation, development and germination. In: Warm season (C4)grasses (eds Moser L.E., Burson B.L. and Sollenberger, L.E.), pp. 95—143. American Society of Agronomy Inc., Madison. Lodge G.M. and McCormick L.H. (2010) Tropical perennial grasses - seed quality. Primcfact 1023. Department of Industry and Investment NSW, Sydney. McKays Grass Seeds (2014) Categories: Buffalo, Couch, Kikuyu, Queensland Blue Couch URL: (accessed 1.3 March 2015). McCormick L.H., Lodge G.M., Boschma S.P. and Murray S. (2009) Simple rules to use when buying seed of tropical perennial grasses. In: Proceedings of the 24th Annual Conference of the Grassland Society of NSW (eds Brouwer D., Griffiths N. and Blackwood I.), pp. 97-100. The Grassland Society of NSW Inc., Orange. MclvorJ.G. and Retd D.J. (2011) Germination characteristics of tropical and sub-tropical rangeland species. The Range/and Journal 33,195-208. Merritt D. (2006) Seed storage and testing. In: Australian seeds. A guide to their collection, identification and biology, (eds Sweedtnan L. and Merritt D.), pp. 53—60. CS1RO Publishing, Canberra. Nolan M., Windsor D. and Williamson S. (1997) Dormancy and viability of Themeda australis (kangaroo grass) and its implications for mining revegetation in die Central Tablelands of New South Wales. In: Proceedings of the Second Australian Workshop on Native Seed Biology for Revegetation. (eds Bellairs S.M. and Osborne J.M.), pp. 149-154. Australian Centre for Minesite Rehabilitation Research, Brisbane. Silcock R.G., Williams L.M. and Smith F.T. (1990) Quality and storage characteristics of the seeds of important native pasture species in south-west Queensland. Australian Rangeland journal 12, 14-20. Wells G.B., Davidson PJ., Dixon K.W. and Adkins S.W. (2000) Defining seed quality of Australian arid zone hummock grasses (Triodia and Plectrachne spp.). In: Proceedings of the Third Australian Workshop on Native Seed Biology for Revegetation. (eds Asher C.J. and Bell L.C.), pp. 59-83. Australian Centre for Minesite Rehabilitation Research, Brisbane. Northern Territory Naturalist (2016) 27: 47-53 Research Article Nest site fidelity of Flatback Turtles (Natator depressus) on Bare Sand Island, Northern Territory, Australia Natalie Bannister, John Holland and Trisia Farrelly Environmental Management Group, Institute of Agriculture and Environment, College of Sciences, Massey University, PMB 11 222, Palmerston North 4442, New Zealand Email: j.d.holland@massey.ac.nz Abstract The endangered Flatback Turtle ( Natator depressus) is endemic to the continental shelf of northern Australia and is the only species of marine turtle with such a restricted geographical distribution. Most mature female Flatback Tunics show a high degree of fidelity to their chosen nesting beach, returning to the same beach within the same and successive nesting seasons. Natal homing has been well studied in other species of marine turtles and our findings support the supposition that all marine turdes display a similar degree of natal homing. Our study area is Bare Sand Island, Northern Territory, where we investigated nest site fidelity of female Flatback Turtles and the influence of wind speed, air and sand temperature, and relative humidity on nest site selection. The data were collected during a 46-day period from 12 June 2012. On Bare Sand Island, female Flatback Turdes demonstrate very strong nest site fidelity, with consecutive nests being located 247 m ± 198 s.d. apart. During the peak 2012 breeding season, sand temperatures, wind speed and relative humidity remained constant, however there was a significant difference in the air temperature between nesting days. Our study of the effects of environmental factors on the nesting environment of Flatback Turtles will contribute towards management practices to protect this endangered species. Introduction The range of the Flatback Turde {Natator depressus) is limited to the continental shelf of northern Australia and its distribution is the most geographically restricted of all marine turde species (Limpus 2009). Unlike other marine turtles, Flatbacks lack an oceanic phase in their life cycle and remain in the surface waters of the continental shelf (Walker & Parmenter 1990). They are identified by four pairs of costal scales on a low'-domcd carapace and a pair of prefrontal scales on the head. The carapace of the hatchling measures approx. 6 cm and weighs approx. 43 g. At maturity, the adult female Flatback Turtle is olive grey, has an average carapace length of 92 cm and weighs around 90 kg (Limpus 2009). Flatback Turdes come ashore to lay their eggs on remote beaches along the tropical and subtropical northern Australian coastline (Whiting et al. 2008). As a consequence of the destruction of nesting habitats and fishing bycatch, the Environmental Protection of Biodiversity and Conservation (EPBC) Act (1999) has classified 48 Northern Territory Naturalist (2016) 27 Bannister et al. the Flatback Turtle as Endangered. The Flatback is the only marine turde that the International Union for Conservation of Nature lists globally as Data Deficient (IUCN 2010). Marine turtles are becoming increasingly threatened as a result of commercial fishing bycatch, boat strike, illegal harvesting of eggs and adults, and an increase in nesting habitat destruction (Limpus 1995; Whiting & Guinea 2003). Of all the species of marine turdes, the Flatback is by far the most under-researched and under-reported (Pendolcy et al 2014). It has been suggested that the breeding site selection of marine turdes may be attributed to individual animals imprinting on magnetic fields in their natal area and then, years later, using this information to return to their natal site (Hueter 1998; Lohmann et al 2008). However, uncertainty remains as to whether their homing is attributable to imprinting to the natal beach as a hatchling, or imprinting to the region of their birth and then the specific beach as an adult during their first breeding season (Limpus 2009). Genetic analyses show that the precision of natal homing can vary considerably among different populations and species with homing to regions of coastline measuring several hundred kilometres being common (Lohmann et al 2008). Female Flatback Turdes generally lay three to four clutches of eggs each nesting season with an inter-nesting interval of approx. 14 days (Hcwavisenthi & Parmenter, 2002). Mature female Flatback Turtles show a high degree of fidelity to their nesting place, returning to the same beach to lay consecutive clutches within a nesting season and in successive nesting seasons (limpus 2009; Oates 2010; Matos et al. 2012). The environment of the nest site is important for marine turdes because it influences offspring sex, embryonic survival, hatchling development rates and hatchling size, mass and energy reserves (Hcwavisenthi & Parmenter 2002; Koch et al 2007). To date, however, there is little understanding of the physical factors that contribute to the distribution of turtle nesting (Santana Garcon et al. 2010). What is known, is that sex determination is temperature-dependent; higher temperatures produce females, whilst lower temperatures result in male hatchlings (Santana Garcon et al. 2010; Hcwavisenthi & Parmenter 2002). The northern tropical Australian populations of Flatback Turdes have a protracted nesting period of around nine months of the year, reaching a peak in July (Whiting et al. 2008). Several studies have attempted to examine the influence of beach characteristics on nesting cycles (Parmenter & Limpus 1995; Hcwavisenthi & Parmenter 2002; Koch et al 2007; Whiting et al 2008). In this study we investigated nest site fidelity of female Flatback Turdes on Bare Sand Island within a single breeding season to assess the influence of environmental factors on nest site selection. The study site was free of introduced predatory species and human interference was minimal. The environmental factors considered wurc wind strength, relative humidity', and air and sand temperatures. Nests of Flatback Turtles Northern Territoiy Naturalist (2016) 27 49 Turtle GPS Points Fig. 1. Location of nest sites where female Flatback Turtles laid more than once during the 2012 peak breeding season on Bare Sand Island, Northern Territory. (Courtesy Digital Globe 2012) Two-hour foot patrols either side of the evening high tide were undertaken to coincide with the turtles’ main nesting activity. Turdes were identified by their tags or were tagged on¬ site and the nest location was recorded using Garmin GPSmap 60CSx, or Garmin GPS72, or Garmin GPS72H. GPS points, and their corresponding tags, were uploaded into PS HI ArcMap 10.1 (© 2012, ESRI, Redlands, California) and a Microsoft® transposed onto a Google Earth Image l.andsat© 2012 Google using the Landsat 7 Satellite (see Fig. 1). The image was geo-referenced against known ladtude and longitude co-ordinates and projected into the WGS 84 Global Positioning System with our accuracy of approximately 5 m. A Hobo™ water temp Pro v2 data logger was buried in the sand on top of a dune at the same depth as that of a Flatback Turtle’s nest (50 cm) and another was positioned in the middle of the western beach. The sand temperature was recorded every half hour and air temperatures and wind speeds were recorded daily at 18.45 hr using a hand-held Air™ speed temperature meter (Dick Smith Qf301). The meteorological data were compared with readings from the Australian Bureau of Meteorology weather station (AWS Station 014277) located on Dum In Mirrie Island (13 km south of Bare Sand Island) and we were satisfied that the Meteorological Station data were suitable for our study purposes. Materials and Methods Our study site is Bare Sand Island (12°32.39’S, 130°25.02’E), which is 50 km west of Darwin, Northern Territory. It is located towards the end of a chain of eight islands (Whiting & Guinea 2003; Koch & Guinea 2006). The main nesting beach faces west and is composed of fine sand with a gentle rise, making the nesting beach easily accessible from the ocean (Koch & Guinea 2006). The data were collected during a six and a half week period from 12 June 2012. 50 Northern Territory Naturalist (2016) 27 Bannister et al. For each female, the distance between consecutive nests during the 2012 season was calculated using Arc Map on-board measuring toolsets. These distances were compared with those from previous nest sites to an equal number of randomly generated potential nest sites and a two sample /-test was used to determine whether the two samples were significantly different. To determine whether there were differences between environmental conditions, paired /-tests were used to compare the environmental data from the different days when female I'latbacks nested. Both the paired and /-tests were analysed using alpha values of 0.05. All the data were analysed using Microsoft Excel 2010 (© 2010 Microsoft Corporation, Redmond, Washington, U.S.A) and Minitab 15 (© 2007 Minitab Inc., State College, Pennsylvania, U.S.A.). Results Fifty-four female Flatback Turtles were observed coming onto Bare Sand Island to nest more than once. Of these, four returned three times, wo nested four times and the rest nested twice. The mean inter-nesting period was 19 days ± 1.58 s.d. with the minimum and maximum inter-nesting periods being 15 days and 37 days, respectively. The locations of nest sites for females that nested more than once on Bare Sand Island are shown in Fig. 1. The mean individual inter-nesting distance was 247 m ± 198 s.d. with a minimum of 7 and a maximum 687 m. We found that more than 50% of the females laid their second nest within 250 metres of their initial nesting site and that they were non randomly selected (/ = -4.79; P - 0.000; d.f. — 108). The mean distance between two nest sites of the same female was 247 m ± 198 s.d., whereas the mean distance between randomly selected sites is 456 m ± 277 s.d.. The mean air temperature on Bare Sand Island was 24.35°C ± 1.25 s.d. and was not significandy different from the meteorological recordings on Dum In Mirrie Island (/ = -0.89; P = 0.376; d.f. — 52). The same was true for wind speeds, with a Bare Sand Island mean of 10.09 km/h ± 5.0 s.d.(/ = 0.36; P = 0.718; d.f. = 42). The mean wind speed was 9.75 km/h ± 2.60 s.d. and the mean relative humidity was 52.1% ± 19.0. The effects of four environmental elements on nesting behaviour were studied; namely, air and sand temperatures, wind speed and relative humidity. The mean values of the four sets of data were recorded on days when the same female turtle nested and were compared to determine whether they differed significandy (see Table 2). Using paired /-tests, we found no significant difference between the sand temperature, wind speed or relative humidity on different nesting days, however there was a difference in air temperatures. Table 2. Comparison of environmental factors influencing nest site conditions of the same female turtle on consecutive nesting times. Mean Nest 1 Mean Nest 2 t- Value P- Value Sand temperature 28.16 t 0.4* 28.17 ± 0.48 °C -0.06 0.950 Wind speed 9.98 ± 2.55 km/h 10.23 + 2.48 km/h -0.52 0.608 Relative humidity 43.07 ± 15.50% 45.25 ± 18.07% -0.69 0.494 Air temperature 24.72 ± 1.41 °C 25.47 ± 1.44 °C -2.57 0.013 Nests of Flatback Turtles Northern Territory Naturalist (2016) 27 51 Discussion The number of Flatback Turtles that returned to Bare Sand Island to nest during the 2012 breeding season was lower than in previous years - a fact that may be attributed to it being the coldest July in 35 years (Australian Government Bureau of Meteorology 2012). The nesting rates of the turdcs observed in our study were generally less than the 3-4 clutches expected for Flatback Turtles during a nesting season (Hcwavisenthi & Parmcnter 2002), however, our data collection period was limited to the peak nesting period. The inter-nesting period we recorded was 19 days ± 1.58 s.d. This is longer than the mean 15-day inter-nesting interval that usually occurs with Flatback Turdes within the same season (Limpus etal 1984; Hewavisenthi & Parmcnter 2002). Again, this might be attributed to the cooler water temperatures that are known to reduce the rate of pre- ovipositional development of eggs during inter-nesting for Loggerhead, Olive Ridley and Green Turtles (Sato et al 1998; Flays el al 2002; Matos et al 2012). From Figure 1 it can be seen that the study animals preferentially nested on the western beach on Bare Sand Island. This is consistent with previous years and is attributed to the fine sand and the gently sloping beach that faces the open ocean (Koch & Guinea 2006). Although it is not entirely clear why some beaches arc preferentially selected by sea turtles to deposit eggs, a number of factors have been identified. The beach must be easily accessible from the ocean, be high enough to avoid inundation at high tide and have temperatures conducive to egg development (Miller etal 2003). The south-easterly section of the island is exposed to strong trade winds throughout the nesting season which explains its low nest density (Koch & Guinea 2006; Koch 2007). The high degree of nest site fidelity observed in our study animals agrees with the similar findings of Limpus (2009), (fates (2010) and Matos etal. (2012). Our study suggests that Flatback eggs may be more tolerant of higher incubation temperatures than those of most other sea turtle species; a finding that is supported by the work of Hewavisenthi & Parmcnter (2002). This change in air temperature may have contributed to the females’ nest site choice, as the nesting phase of the marine turdes’ reproductive cycle is thought to be largely determined by temperature (Santana Garcon et al 2010). Fligher temperatures will reduce the turdes’ progress across a beach to the ocean. However, as they typically emerge at night, their movements are not hindered by high temperatures (Koch et al 2008). The mean inter-nesting distance of 247 ± 198 m for Flatback Turtles differs markedly from Olive Ridley Turdcs (4.83 ± 4.37 km) and Green Turdes (0-5 km) (Matos et al 2012; Lalith Ekanayakc et al 2003). Sand temperature, however, plays a vital role in the development of turdes and influences hatchling size, sex, and energy reserves and successful incubation is only possible within certain thermal limits. Nest temperatures are variable not only on a single beach within a season and at different levels on the shore, but also vary with depth at a single nest site (deeper eggs are incubated at rather lower, more stable temperatures). The influence of short periods of extreme temperature is unclear. However, it has been reported that 52 Northern Territory Naturalist (2016) 27 Bannister et al the final third of the incubation period is particularly temperature-sensitive and eggs rarely hatch if exposed to temperatures below 23°C or above 33°C (Davenport 1997). Flatback Turdes appear to be more tolerant of high incubation temperatures and severe moisture stress than most marine turtle species (Hewavisenthi & Parmenter, 2002). Sand temperature influences the timing of the emergence of Flatback hatchlings. Most left the nest during the same few hours each night because of thermal cues that are dependent upon a combination of threshold temperatures, thermal gradients in the nest, and rates of temperature change (Davenport 1997; Koch et al. 2008). As sand temperatures at the study site remained below 29.3°C (the pivotal temperature) and as sex determination is temperature-dependent, predominantly male hatchlings emerged (Koch et at. 2007). Strong south-easterly seasonal trade winds have persisted throughout the nesting season on Bare Sand Island over the last 10 years and, while this has the potendal to affect the depth of the nests, they were not reported to have been shallower than 20 cm and were not subsequently threatened by the high temperatures that occur late in the nesting season (Koch & Guinea 2006; Koch et a1 2007). We found no significant difference between the mean relative humidity (52.1% ± 19.0 s.d.) on successive nesting days. Favourable conditions for embryonic development and survival include high humidity influencing hatchling size, sex, and energy reserves (Hewavisenthi & Parmenter 2002; Miller et al. 2003). References Australian Government Bureau of Meteorology (2012) Monthly climate summaryfor Northern Territory. Northern Territory in July: cool dry season conditions, (accessed 30 September 2013) Booth D.T. and Astill K. (2001) Temperature variation within and between nests of the Green Sea Turtle, Chelonia mydas (Chclonia: Cheloniidac) of Heron Island, Great Barrier Reef. Australian Journal of Zoology 49, 71-84. Davenport J. (1997) Temperature and the life-history strategics of sea turtles. Journal of Thermal biology 22, 479-488. Hays G.C., Broderick A.C., Glen F., Godley B.J., Houghton J.D.R. and Metcalfe J.D. (2002) Water temperature and internesting intervals for Loggerhead ( Caretta caretta) and Green ( Chelonia mydas) Sea Turtles. Journal of Thermal biology 27, 429-432. 1 lewavisenthi S. and Parmenter C.J. (2002) The incubation environment and nest success of the flatback turtle (Nata/or depressus) from a natural nesting beach. Coptia 2002,302—312. I lueter R.E. (1998) Philopatry, natal homing and localised stock depletion in sharks. Shark. News: Newsletter of the IVCN Shark Specialist Group 12, 1—2. IUCN (2010) IUCN Red List of threatened species. Version 2013.2 (accessed 30 September 2013) Koch A.U. and Guinea M.L. (2006) Lower nesting success of flatback turtles caused by disorientation. Marine Turtle Newsletter 114, 16. Koch A.U., Guinea M.L., and Whiting S.D. (2007) Effects of sand erosion and current harvest practices on incubation of the Flatback Sea Turtle ( Natator depressus). Australian journal of Zoology 55, 97-105. Nests of Flatback Turtles Northern Territory Naturalist (2016) 27 53 Koch A.U., Guinea M.L. and Whiting S.D. (2008) Asynchronous emergence of Flatback Sea Turtles, Natator depressus, from a beach hatchery in Northern Australia, journal of Herpetology 42, 1-8. Lalith Ekanayake E.M., Ranawana K.B. and Kapurusinghe T. (2003) Nest site fidelity of Green Turtles on the Rekawa Turtle Rookery in Southern Sri Lanka. In: Proceedings of the 23rdAnnual Symposium on Sea Turtle Biology and Conservation. NOAA Technical Memorandum NMFS- SEFSC-536, USA (comp. Pilcher N.J.), p. 159. Limpus C.J. (1995) Conservation of marine turtles in the Indo-Pacific Region. Australian Nature Conservation Agency, Australia. Limpus C.J. (2009) Flatback turtle, Natator depressus (Garman). In: A biological review ofAustralian marine turtles, (ed. Limpus C.J.), pp. 246-298. Queensland Environmental Protection Agency, Brisbane. Limpus C.J., Fleay A. and Baker V. (1984) The flatback turtle, Chelonia depressa, in Queensland: reproductive periodicity, philopatry and recruitment. Australian Wildlife Research 11, 579-587. Lohmann K.J., Putman N.F. and Lohmann C.M.F. (2008) Geomagnetic imprinting: A unifying hypothesis of long-distance natal homing in salmon and sea turtles. Proceedings of the National Academy of Sciences 105, 19096-19101. Matos I.., Silva A., Castilhos J., Weber M., Soares L. and Vicente L. (2012) Strong site fidelity and longer internesting interval for solitary nesting Olive Ridley Sea Turtles in Brazil. Marine Biology 159,1011-1019. Miller J.D., Limpus C.J. and Godfrey M.H, (2003) Nest site selection, oviposition, eggs, development, hatching and emergence of Loggerhead Turtles. In: Loggerhead Sea 'Turtles (eds Bolten A.B. and Witherington B.E.), pp. 125-143. Smithsonian Books, USA. Oates J. (2010) Flatback turtle tagging program at Cemetery Beach 2009/2010. Pendolcy Environmental Pty Ltd., Australia. Parmenter C.J. and Limpus C.J. (1995) Female recruitment, reproductive longevity and inferred hatchling survivorship for the flatback turtle ( Natator depressus) at a major eastern Australian rookery. Copeia 1995, 474—477. Pendoley K., Bell, G, McCracken R., Ball K., Sherborne J., Oates J., Becker P, Vitenbergs A. and Whirtock P. (2014) Reproductive biology of the Flatback turtle Natator depressus in Western Australia, Endangered Species Research 23 (2014), 115-123. Santana Garcon J., Grech A., Moloney J. and Hamann M. (2010) Relative exposure index: an important factor in sea turdc nesting distribution. Aquatic Conservation: Marine and Freshwater Ecosystems 20, 140—149. Sato K., Matsuzawa Y., Tanaka LL, Bando T., Minamikawa S., Sakamoto W. and Naito Y. (1998) Internesting intervals for Loggerhead Turdes, Caretta caretta, and Green Turdes, Chelonia mydas, are affected by temperature. Canadian journal of Zoology 36, 1651-1662. Walker T.A. and Parmenter C.J. (1990) Absence of a pelagic phase in the life cycle of the flatback turde, Natator depressa (Garman). journal of Biogeograplry 17,275—278. Whiting A.U., Thomson A., Chaloupka M. and Limpus C.J. (2008) Seasonality, abundance and breeding biology of one of the largest populations of nesting flatback turdes, Natator depressus. Cape Domett, Western Australia. Australian Journal of Zoology 56, 297-303. Whiting S.D. and Guinea M.L. (2003) The nesting biology of flatback turtles in the tropics: Seven years of surveys on Bare Sand Island, Darwin, Northern Territory, Australia. In: Proceedings of the 23 rd Annual Symposium on Sea Turtle Biology and Conservation. NOAA 'Technical Memorandum NMFS-SEFSC-536, USA (comp. Pilcher N.J.), p. 159. 54 Northern Territory Naturalist (2016) 27: 54-59 Short Note Records of waterbirds and other water-associated birds from the 2014/15 migratory season in the Darwin region Amanda Lilleyman Research Institute for the Environment and Livelihoods, Charles Darwin University, Darwin, NT 0909, Australia Email: amanda.lilleyman@cdu.edu.au Abstract Records of waterbirds, waterfowl, terns, gulls, egrets and herons, raptors, and resident shorebirds in the Darwin region. Northern Territory, were collected during fortnighdy migratory shorebird monitoring. Eight study sites were monitored from August 2014 through to April 2015, which is considered the migratory season for most non-passerine birds in the Top End. Species abundance across the sites, breeding records, and new information on habitat use at an artificial habitat (East Arm Wharf) are presented. Across the eight study sites there were 39 species recorded, representing 15 taxonomic families. Darwin Harbour in the Northern Territory has a rich coastal waterbird assemblage, owing to its diverse range of habitats. The coastal region supports resident and nomadic Australian waterbirds, waterfowl, resident and migratory terns and gulls and various raptors that inhabit coastlines (McCric & Watson 2003). A number of terns that breed in the northern hemisphere visit northern Australian coastlines during the austral summer where they feed over the ocean and along tide lines and then roost at beaches, rocky reefs, dykes and on floating buoys. The macro-tidal nature of tides in the Darwin region creates extensive mud and sand flats, available for foraging birds. Mangroves, saltpans and saltmarsh provide roosts duting high tides. Studies of waterbirds in the Top End have mainly focused on freshwater wetlands and floodplains in the Fogg Dam and Alligator River regions, east of Darwin (sec Crawford 1979; Morton et al. 1993). The waterbirds in Darwin Harbour prefer coastal saline habitats, including fringing mangroves, brackish waste water ponds and dredge ponds nearby, and creeks and rivers. Extensive aerial and ground surveys along the Northern Territory coastline indicate that the region supports a variety of waterbirds (Chatto 2006). During regular monitoring of migratory shorebirds I collected count data for all birds across eight study sites from August 2014 through to April 2015, which is when most migratory shorebirds and other water-associated migrant birds visit Australian shores. Birds were surveyed at each site most fortnights during spring tides, which were selected ^aterbirds in the Darwin region Northern Territory Naturalist (2016) 27 55 to target when migratory shorebirds would be roosting. There were 184 surveys performed over the nine survey months. The sites were East Arm Wharf, Lee Point- Buffalo Creek, Ludmilla Bay, Spot On Marine boat yard, Nightcliff Rocks, East Point, Sandy Creek and East Arm Wharf Railway Mud, all within the Darwin region. This note summarises the results of all birds present at die study sites, excluding migratory shorebirds. Thirty-nine species of birds were recorded within the study period, including 5 species of heron and egret, 2 gull, 8 tern, 3 raptor, 8 resident shorebird, 8 waterbird and 5 waterfowl species. The maximum count for each species and the corresponding site and date are shown in Table 1. Table 1. Results from waterbird monitoring in the Darwin region from August 2014-April 2015. Bird species are grouped and presented in taxonomic order following Chrisridis & Boles (2008). Family and grouping Common name Scientific name Max. count Site of max. count Site co-ordinates Date of max count Waterfowl Anatidae Wandering Whistling Duck Dendrocygna arcuata 149 East Arm Wharf 12.5325°S, 131.0639°E 4 Jan 2015 Anatidac Radjah Shelduck Tadorna radjah 200 lee Point-Buffalo (>cek 12.3453°S. 130.9825°E 21 Nov 2014 Anatidae Pacific Black Duck Anas superciiiosa 17 East Arm Wharf 12.5325°S, 131.0639°! 23 Dec 2014 Anatidae 1 lard head Aytfrya australis 12 East Arm Wharf 12.5325°S. 131.0639°E 6 Apr 2015 Podicipedidae Australasian Grebe Taclrybaptus noiHiehollandiae 12 East Arm Wharf 12.5325°S, 131.0639°E 20 Apr 2015 Waterbirds Anhingidac Australasian Darter Anhirtga novaebollandiae 1 East Arm Wharf 12.5325°S, 131.0639° E. 7 Sep 2014 Phalacrocoracidae Little Pied Cormorant Microcarbo melanaleucos 17 East Arm Wharf 12.5325°S, 131.0639° E 21 Jan 2015 phalacrocoracidae I .ittle Black Cormorant Phalacroiorax su/arostris 5 [•last Arm Wharf Railway Mud 12.6819°S, 130.9800° K 7 Oct 2014 Phalacrocoracidae Pied Cormorant Phalacrocorax mrius 3 Lee Point-Buffalo Creek 12.3453°S, 130.9825°E 21 Nov 2014 Pclecanidae Australian Pelican Pelecanr/s conspicjllatus 44 East Arm Wharf 12.5325°S, 131.0639° E 4 Jan 2015 Ciconiidae Black-necked Stork liphippiorfrynchus asiaticus 2 lAasr Arm Wharf 12.5325°S, 131.0639°E 23 Dec 2014 Egrets and Herons Ardeidae Great Egret Ardea modest a 11 East Arm Wharf Railway Mud 12.6819°S. 130.9800° K 7 Oct 2014 Ardcidac Striated 1 leron Butohdes striata 5 East Arm Wharf Railway Mud 12.6819'S, 130.9800°!': 7 Oct 2014 Ardeidae Pied Heron Egre/fa picata 1 Lee Point-Buffalo (’reek 12.3453°S, 130.9825°E 18 Mar 2015 Ardeidae Little Egret Egretta gar~etta 10 lee Point-Buffalo Creek 12,3453°S, 130.9825T. 12 Aug 2014 Ardeidae Eastern Reef Egretta sacra 12 East Arm Wharf Railway Mud 12.6819°S, 130.9800°E 7 Oct 2014 Waterbirds rhreskiornithidac Australian White Ibis Threskiornis molucca 11 East Arm W'harf Railway Mud 12.6819°S, 130.9800° E 7 Oct 2014 Continued on next page 56 Northern Territory Naturalist (2016) 27: xx-xx Lilleyman Continued from previous page Family and grouping Common name Scientific name Max. count Site of max. count Site co-ordinates Date of max. count Threskiomithidae Royal Spoonbill Plataka rega 13 Fast Arm Wharf I2.5325°S, 131.063913 22 Nov 2014 Birds of Prey Accipitridae White-bellied Sea-1 £agle Haliaeetus kucogtsttr 4 East Arm Wharf 12.5325°S, 131.0639°H 23 Dec 2014 Accipitridac Whistling Kite Haliastur spherturus 1 Hast Arm Wharf 12.53251, 131.06391-' 20 Apr 2015 Accipitridae Brahminy Kite Haliastur tndus 2 East Arm Wharf 12.5325°S, 131.0639°E 23 Dec 2014 Resident shorebirds Burhinidae Beach Stonc- curlcw Esacus ma^nirostns 3 Sandy Creek 12.49441, 130.89061; 23 Mar 2015 1 laematopodidae Australian Pied (hstercatcher Haematopus longrostris 5 Hast Arm Wharf 12.53251, 131.063913 13 Aug 2014 f lacmatopodidae Sooty ( )vstercarcher Haematopus fuliginosHS 5 Hast Point 12.41171, 131.059213 4 l*'cb 2015 Rccurvirostridac Black, winged Stilt / limantopus himantopus 74 Hast Arm Wharf 12.53251, 131.063913 20 Apt 2015 Charadriidae Red-capped Plover Cbaradrius rufiaipiUus 16 I.cc Point-Buffalo Creek 12.34531, 130.982513 19 Dec 2014 Charadriidae Red-kneed Dotterel Erythrogonys cinctus 2 Hast Arm Wharf 12.53251, 131.063913 20 Apr 2015 Charadriidae Masked 1 .apwing Vanellus miles 20 Spot On Marine 12.64111, 130.88641- 8 Mar 2015 GlarcoLidac Australian Pratincole Stiltia isabelJa 2 Hast Arm Wharf 12.53251, 131.06391 20 Apr 2015 Terns and Gulls I .andae 1 .ittlc Tern Stemula albifrons 29 1 x*e Point-Buffalo (Teek 12.3453°S, 130.9825° H 24 Mar 2015 I .andae Gull-billed Tern (macrotarsd ) 1 Gelochelidon nilotua mairotarsa 234 Hast Arm Wharf 12.5325°S, 131.0639® E 7 Sep 2014 I.aridae Gull-billed Tern (affinis)' Gelochelidon nilotica affims 6 Hast Arm Wharf 12.53251, 131.06391- 2 Nov 2014 I .aridae Caspian Tern f [ydroprogne caspta 6 Sandy Creek 12.4944°S, 130.8906° H 23 Mar 2015 Laridae Whiskered Tern Chlidonias hyhrida 351 Hast Arm Wharf 12.53251, 131.06391- 23/12/2014 Laridae White-winged Black Tern Cblidonias leucopterus 274 Hast Arm Wharf 12.53251, 131.06391- 23/12/2014 Laridae Common Tern Sterna hirundo 1 East Point 12.41171, 131.05921 12/10/2014 1 .aridae I x*sser Crested Tern Thalasseus bengalensis 35 l-ee Point-Buffalo Creek 12.3453°S, 130.9825° H 23/11/2014 laridae Crested Tern Thalasseus bergn 192 Nightcliff Rocks 12.59811, 130.9531°K 9/11/2014 Laridae Franklin's Gull Eeucophaeus pipixean 1 late Point Buffalo < 'stock 12.34531, 130.98251'. 18 Mar 2015 1 .aridae Silver Gull Chroicocephalus not'oehollandiae 480 I jcc Point-Buffali> Creek 12.34531, 130.98251 18 Mar 2015 'Two subspecies of Gull-billed Tern occur in northern Australia, affinis being a migrant that visits Australia during the summer season. These subspecies can be separated in the field using morphological features; see Lilleyman and Hcnsen (2014). East Arm Wharf, an artificial site made up of dredge ponds, situated within Darwin Harbour, consistently supported the most species of all the sites. The site attracts a diverse range of species because the ponds represent a mixture of freshwater and Waterbirds in the Darwin region Northern Territory Naturalist (2016) 27 57 marine/saline habitats, with input from the harbour. The ponds are in open terrain with good visibility for birds to detect predators, and situated next to the coastline. The site is also protected from human disturbance as public access is restricted, and the site excludes feral terrestrial predators like dogs and cats through trapping and fencing. East Arm Wharf supported the most species (18) of water-associated bird compared to the other sites during the monitoring period. Sixteen species were recorded at the mm Fig. 1. Franklin’s Gull (right) and Silver Gull (left) in a dredge pond at East Arm Wharf in Darwin, 20 April 2015. (Amanda Lilleyman) East /Arm Wharf Railway Mud (adjacent to the dredge ponds at East /Arm Wharf), but the assemblages between these two close sites varied. Twelve species of waterbird were recorded at Lee Point-Buffalo Creek during the monitoring period. Across the sites, the month of November had the highest total count of birds, mosdy weighted by terns, followed by March and then December, both weighted by gulls and terns. A vagrant gull, Franklin’s Clull (/ jeucophaeus pipixean) (Fig. 1), distinguished from the more common Silver Cull {Choicepbalus novaeboHandiae) by its black head markings or prominent hood and dark grey back and upperwings contrasting with white underparts, was recorded in March, initially at Buffalo Creek, and subsequently at Stokes I lill Wharf (Mark de Krctser, pers. comm. 18 April 2015), and lastly at East Arm Wharf (by AL). This species breeds in North America and spends the non-breeding season in South America ( Handbook of the Birds of the World Alive 2015). This is the 19th record for Australia and the second time the species has been recorded in the Northern Territory (BirdLife Australia 2015). The first arrival and last departure records for migratory terns and one vagrant gull are shown in Table 2. 58 Northern Territory Naturalist (2016) 27 Lillcyman Crawford (1980) reported mean counts Table 2. First arrival and last departure records for Whiskered Terns (Cbltdonias hybrida) for migratory terns and one vagrant gull Bird t ^ species are presented by their first arrival month, with peaks in September at Fogg Dam and in Darwin (100 and 50 individuals, respectively) and Lesser Crested Terns (Thalasseus bengalensis ) (40 individuals) along the coastline of Darwin, but noted in January for the highest mean counts of Crested Terns ( Thalasseus bergii) (100 individuals). The maximum count of Crested Terns from the current study was 192 individuals in November from Nightcliff, which was a site not surveyed in the Crawford (1980) paper. The maximum count of Whiskered Terns from the current study is certainly an increase from the mean counts recorded by (Crawford 1980). Outside the monitoring period, in June, July and August there were up to 10 Red-necked Avoccts ( Recurvirostra uovaeboHaudiae) using the freshwater dredge ponds at East Arm Wharf. Records from eBird and a local online forum (NT Birds Yahoo group) show this species is recorded in the Top End every year or so, but mostly further south and east of Darwin city (i.e. South Alligator River, Mamukala, Shark Billabong, Adelaide River). This record at East Arm Wharf is the first record close to Darwin since 20 August 2013, when the species was recorded at Leanyer Sewage Treatment Ponds. Resident shorebirds were recorded nesting and raising young at East /Arm Wharf, including Black-winged Stilts {Himantopus himantopus), with a maximum count of 74 individuals in April, after the nesting period. Red-capped Plovers {Charadrius ruficapillus), Masked Lapwings {VaneHus Species First arrival Last departure Gull-billed Tern (. affinis ) early ()ctobcr February (Common Tern tnid-( )ctobcr March White-winged Black Tern late October late April I attic Tern late December early April Franklins Gull March April Fig. 2. Red-necked Avocets in a dredge pond at East Arm Wharf in Darwin, 19 July 2015. (Amanda Lilleyman) Waterbirds in the Darwin region Northern Territory Naturalist (2016) 27 59 miles) and Pied Oystercatchers (Haematopus longirostris) were also recorded breeding along the muddy edge of one of the dredge ponds. Red-capped Plover also regularly breeds along the sandy beach at Lee Point. (Other localities around Darwin provide quality habitat for waterbirds, shorebirds and other water-associated birds, including Holmes Jungle, Knuckey Lagoon, McMinns Lagoon, Leanycr and Palmerston Sewage Treatment Ponds. These sites were not surveyed in the study period as they were not included in the migratory shorebird monitoring program; however, future monitoring of these sites would improve our knowledge of birds in the region. The Darwin region coasdine and associated freshwater ponds support a diverse range of water-associated bird species and high abundances throughout the austral summer season. Of the sites surveyed for this study. East Arm Wharf is the most important site (based on species diversity and number of individuals recorded) for a range of waterbirds, terns and gulls, waterfowl and breeding resident shorebirds. Acknowledgements Thanks to Darwin Port Corporation for allowing the author ongoing access to East Arm Wharf. Thanks to Bas Henscn for providing helpful comments and feedback on this paper. Thank you to the anonymous reviewer who provided comments on this manuscript and to Richard Willan for his editorial comments. References BirdLife Australia (2015) Rarities Committee. Birdl.ifc Australia, Melbourne, Victoria. (accessed 21 September 2015). Chatto R. (2006) The distribution and status of waterbirds around the mast and coastal wetlands of the Northern Territory. Parks and Wildlife Commission of the Northern Territory. Christidis L. and Boles W.E. (2008) Systematise and Taxonomy of Australian Birds. CSIRO Publishing, Collingwood. Crawford D. (1979) Waterbirds: indices and fluctuations in dry-season refuge areas, Northern Territory. Wildlife Research 6, 97-103. Crawford D. (1980) Seasonal fluctuations in numbers of three species of tern in Northern Territory. Emu 80,166—169. I landbook of the Birds of the World Alive (2015) Franklins Gull (lairuspipixean). Lynx Edicions, Barcelona, Spain, (accessed 21 September 2015). Lillcyman A. and Hensetl B.J. (2014) The occurrence of the Asian subspecies of the Gull-billed Tern (Geiocheiidon nilotica affinis) in the Darwin region, Northern Territory. Northern Territory Naturalist 25, 12-17. McCric N. and Watson J. (2003) / '/riding birds in Darwin, Kakadu and the Top End, Northern Territory, Australia. N. McCric, Casuarina, Northern Territory. Morton S.R., Brennan K. and Armstrong M. (1993) Distribution and abundance of grebes, pelicans, darters, cormorants, rails and terns in the Alligator Rivers Region, Northern Territory. Wildlife Research 20,203—217. 60 Northern Territory Naturalist (2016) 27: 60-67 Short Note Fluctuations in use of urban roost and foraging sites in Darwin by Pied Herons (Ardea picata) John Rawsthorne 10 Macartney Street, Fannie Bay, NT 0820, Australia Email: kim iohn@bigpond.net.au Abstract Pied Herons (Ardea picata ) arc a common bird of Australia’s north, strongly associated with shallow freshwater wetland and estuarine habitats. However they also use urban sites for foraging, and in particular many feed at the Lcanyer Sewage Treatment Works and the Shoal Bay Waste Depot site in Darwin’s northern suburbs. Pied Herons roosted on Catalina Island in Darwin Harbour from at least 2010 to 2013. Seasonal fluctuations in numbers and use of that, and nearby roost sites, are documented here, as is the more recent abandonment of harbour roost sites in favour of constructed riverside habitat at Crocodylus Park, close to the foraging sites. Introduction Pied Herons (Ardea picata) (Fig. 1) are a common bird of Australia’s north, associated with shallow freshwater wetland and estuarine habitats (McKilligan 2005). However they Fig. 1. An adult Pied Heron (Ardeapicata) at Knuckcy Lagoon, Darwin, showing nuptial plumes on the head just prior to the wet season. (John Rawsthorne) Habitat use by Pied Herons Northern Territory Naturalist (2016) 27 61 also use urban sites for foraging, and in particular a large number arrive each day to feed at Leanyer Sewage Treatment Works and the Shoal Bay Waste Depot site in Darwin’s northern suburbs (McCrie & Noske 2015;JRpers. obs.). I was initially intrigued by large early morning flocks flying north over Darwin Railway Station at East Arm in 2010, and a suggestion by Richard Noske that the north- south daily movements were a well-established pattern by then. Further observations of evening flocks flying south over Kormilda College, Berrimah, during 2012 and a chance observation of Pied Herons arriving from the north at dusk at Catalina Island in East Arm, Darwin Harbour were the catalyst for a more formal investigation of these movement patterns. Here I document roosting places of these urban birds, their flight path - including potential conflict with Darwin Airport flight paths - and seasonal fluctuations in numbers. Methods Bird activity sites Catalina Island (12.4900°S, 130.9070°E) is a small (50 x 200 m, 1 ha) island in East Arm, Darwin Harbour. In addition to terrestrial trees including one tall Peanut Tree (Sterculia quadrifolia ) (Richard Willan pers. comm.), it has fringing mangroves including Grey Mangrove (Avicennia marina ) and Mangrove Apple (Sonneratia alba), which approximately double the vegetated area to around 2 ha. At low tide, a sand spit and rocky areas are exposed around the island. Shoal Bay Waste Depot and the Leanyer Sewage Treatment Works are on the north-east fringe of suburban Darwin, about 12 km and 14 km, respectively, north of Catalina Island. Crocodylus Park is close to the waste depot and sewage works, about 3 km south of the waste depot. Evening roost surveys Pied Herons were counted flying from the north to evening roosts in East Arm (Fig. 2) monthly from November 2012 to October 2013. Counts were performed from either East Arm Boat Ramp (about 0.7 km north-cast of Catalina Island) or Berrimah Road near the Vopak Fuel Depot (about 1.5 km north of Catalina Island), depending on wind conditions and roost site being used. In September 2015 a follow-up count was performed at Crocodylus Park. Birds generally arrived to roost in flocks, mostly of less than 100 birds, but occasionally in much larger flocks of several hundred individuals. Counts were of individual birds where flocks were less than approx. 50 in number, but for larger flocks or for those in quick succession, estimates of the number of birds were made based on ‘blocking up’ from smaller counts. 62 Northern Terri tor} 1 Naturalist (2016) 27 Rawsthorne Fig. 2. Satellite image of Darwin area showing Pied Heron daytime feeding sites in the north, Crocodylus Park roost site in mid-image and Catalina Island roost site near Fast Arm Wharf in the south. 1 he large white arrow shows the general evening flight path of birds flying to harbour roosts from their daytime feeding sites. Left inset: Catalina Island showing terrestrial vegetation and mangroves. Right inset: Crocodylus Park artificial river roosting site (prior to current vegetation growth). (Images via Google Maps) Results In November 2012 several preliminary counts and tests of counts were made (4357 individuals on 1 November and 5895 individuals on 8 November). These counts were treated as training exercises to refine the counting procedure. Some tests were made ot my counts of larger flocks against photographs of those same flocks, confirming the general accuracy of my counts. On 22 November, two independent counts were made by Gavin O’Brien and me, producing counts of 5138 individuals (GO’B) and 4953 individuals (|R) arriving to roost on that evening. These tests confirmed the broad accuracy of the counts, and suggested that they might be regarded as being ± 5%, rather than precise counts of individuals. In the data that follow, 1 have used the average of the two 22 November counts (5045 birds) as the data point for November 2012. The number of birds counted in individual surveys (big. 3) ranged from a high of 5895 individuals (preliminary count, November 2012) to a low of 1705 individuals (February 2013). The average number of birds by count across all months in 2012/13 was 2771 individuals. Roost counts were conducted for 12 months, from November 2012 to October 2013. In the first few months of the survey, birds exclusively flew to Catalina Island to roost, and were easily counted flying dirccdy overhead of the East Arm Boat Ramp (Fig. 2). In later months birds also sometimes appeared to roost on South Shell Island, a smaller treeless island about 2 km west-south-west of Catalina Island, while in some evening Habitat use by Pied Herons Northern Territory Naturalist (2016) 27 63 Fig. 3. Monthly count of Pied Herons {Ardeapicata) arriving from the north to roost in Darwin 11 arbour, November 2012-October 2013. There is a notable peak in the early wet season, with lowest numbers in the late wet season. surveys birds flew further west beyond the islands in East Arm, and at least as far as the mangroves fringing Wickham Point. Follow-up observations in September 2015 indicated that Pied Herons had ceased to roost in East Arm, and were now roosting much closer to die sewage ponds and waste depot, at an artificial river at Crocodylus Park on the corner of Vandcrlin Drive and McMillans Road. Construction of this habitat commenced in 2007 and was finalised in May 2014. A one-off count on 28 September 2015 counted approximately 3300 Pied Herons roosting at this site. Flight paths All birds arriving to evening roosts arrived from the north. Observations of mid-flight paths for flocks indicated a c|uitc direct flight path between the Shoal Bay Waste Depot and East Arm, with no intermediate stops and no birds coming on to the flight path any further south than about McMillans Road (Fig. 2). Evening and morning observations in the northern suburbs did not detect birds arriving to or leaving from the northern sites in any direction apart from the south. I did not conduct any night-time surveys at the day time feeding sites. The actual flight padis used by commuting Pied Herons were within about 1 km of Bcrrimah Road. Depending on the wind direction, birds deviated in their evening southward flight slightly east or west of Berrimah Road, but corrected their path as they came closer to roost. The flying height of birds was measured using binocular focus distances. 1 focused on birds flying direedy overhead, and then measured the distance along the ground to the same focus distance. Birds typically flew at a height of about 64 Northern Territory Naturalist (2016) 27 Rawsthornc 50-70 m above me at observation points, but appeared to be flying higher, up to 100 m high, at the mid-point of their commute. The flight path of birds travelling to Crocodylus Park was directly south of the waste depot. Some birds had a direct path over Holmes Jungle to Crocodylus Park. However most birds appeared to follow Vandcrlin Drive, with a disunct left-hand turn shordy before McMillans Road, for the short distance across to the roost site at Crocodylus Park. Discussion All the Pied Herons using the East Arm roost came from the Shoal Bay/Leanyer/ Holmes Jungle area of Darwin’s northern suburbs. I cannot be sure that some birds did not also roost at the feeding sites or at other roosts such as the later-identified roost at Crocodylus Park, although the large numbers roosting at Crocodylus Park is a post-2014 phenomenon (Grahame Webb pen. comm). Thus, my urban population estimates of 1700-5900 individuals should be regarded as minimum counts. Chatto (2000) estimated the Pied Heron breeding population in the Top End to be over 22,600 individuals based on aerial surveys of breeding colonies, with the majority of breeding occurring in the north-western part of the Top End, within about 300 km of Darwin. Morton et aL (1993) estimated that the maximum Pied Heron population in the Alligator Rivers region alone was around 50,000 individuals. The peak population of Pied Herons in Darwin of more than 5000 individuals in the late dry season represents a moderate fraction of the overall Top End population. The food available at the sewage ponds and waste depot appears to be an important resource for a small but significant proportion of the Top End population of this species during the late dry season, and may be maintaining the overall population tn the north-west Top End at a slightly higher level than would otherwise be possible. Pied Herons arc present, but do not breed, in the Darwin urban environment throughout the year. The fewer birds remaining during the late wet season and early dry season correspond with the identified active penod of breeding colonies from January to May (Chatto 2000), suggesting that the majority of Pied Herons leave the urban area for breeding sites m the wet season. The closest breeding colonies - within 100 km of Darwin - are in the mangrove-lined mouths of the Finniss and Adelaide rivers The largest colony observed by Chatto was about 3000 individuals, while the average colony s™ around 1000 individuals (Chatto 2000). The drop in urban population from the - ate c ry season peak to die 2013 breeding season low of over 3000 birds is larger arLlmmL°d l"" ^ C ° 1<5ny ’ thuS m ° St likd >' comprising individuals that a nvc trom and depart to more than one breeding colony. An alternative to local movements is that the significant increase in the number of Pied n, S m DarW ‘ nS u ^ ban s,tes 111 th c early wet season is made up of migratory birds ecendy returned to Australia from the north where they had migrated to escape the Habitat use by Pied Herons Northern Territory Naturalist (2016) 27 65 food and habitat shortages of the late dry season. Pied Herons are known to migrate regularly across the Torres Strait between North Queensland and Papua New Guinea for the dry season (Garnett & Bredl 1985; Nlarchant & Higgins 1990; McKilligan 2005). Although they are present in Papua New Guinea year-round, they are not known to breed there (Coates 1985; Beehler et al. 1986). 1 am not aware of any records of Pied Herons departing from or arriving to the Top I : .nd from the north, but migration of the Top End populations to Papua New Guinea or Indonesia should not be ruled out and is worthy of further investigation. The seasonal cycle of movements of Pied Herons in the Darwin region is not as clear as for other Top End systems. For example, a study on the Magcla Creek floodplain in April 1981 (i.e. towards the end of the Pied Heron breeding season) identified that over 90% of individual Pied Herons observed were in immature plumage (Recher et al 1983; see also Garnett 1985). This is consistent with a broader seasonal pattern in the Magela Creek system of near-complete absence during the wet season and a gradual increase over the dry season, with corresponding later offsetting pulses of birds in different nearby systems linked to wetting and drying cycles of the different floodplains (Morton et al. 1993). Most likely, the April 1981 birds were newly fledged from one of the now- identified breeding colonies on the East or South Alligator rivers (Chatto 2000), and they are gradually joined on the floodplains by adults as they depart the heronries each May. The seasonal offsetting patterns identified by Morton et al. (1993) weakly suggest that birds from that area do not migrate north to Papua New Guinea or Indonesia, but are able to find suitable habitat in the late dry season within the mosaic of drying wetlands in the western Arnhem Land region. There is a pool of non-breeding Pied Herons in Darwin through each wet season that do not congregate in heronries, and there may be others scattered across a wide area of wet floodplain or other habitat. This would explain the larger population estimate of 50,000 birds in the Alligator Rivers region by Morton et al. (1993) compared to Chatto’s (2000) estimate of 22,600 birds present at all breeding colonies in the western Top End. Pied Heron plumage varies by age, with juvenile and immature birds having different crown feather colour and other more subde differences to adults. HANZAB indicates a juvenile plumage and two immature plumages (Marchant & Higgins 1990), suggesting that Pied Herons most likely do not breed until at least the end of their second year of life. Pied Herons arc never completely absent from the urban area of Darwin, unlike other waterbirds such as Magpie Geese that visit the urban area for limited periods and then have seasonal absences, and it may be that the remaining (approx. 2000) herons are immature. As a starting point, closer observation of the plumage of individuals foraging in Darwin at different parts of the seasonal cycle may shed further light on population and breeding dynamics An alternative explanation for the presence of Pied Herons in Darwin during the breeding season may be that the food supply in Darwin in the late dry season is limiting and that the urban area acts as an ecological trap (sensu Robertson & Hutto, 2007) from 66 Northern Territory Naturalist (2016) 27 Rawsthornc which some birds struggle to escape. Pied Herons are regularly trapped accidentally at Crocodylus Park within food preparation areas, and are often noted to be either very skinny or otherwise injured or in poor condition (Simon Ferguson, pers. comm.). In addition to studying the plumage of birds through the seasons, observation of body condition of individuals roosting at Crocodylus Park at different parts of the seasonal cycle may provide insights into source/sink dynamics for the Top End population. Flight path conflict with Darwin airport There is some potential for conflict between Pied Herons when they are flying to or from the harbour roost sites and the short final approach of aircraft arriving at Darwin International Airport from the cast. Given the large size of flocks identified in this study, the potential for multiple strikes exists, although the most common flight path of flocks of Pied Herons appeared to be safely lower than the approach paths of aircraft arriving from the east. Planes taking off to the east appear to climb steeply after take-off, and birds appeared much less likely to conflict with departing planes. Although Darwin airport has a high number of recorded bird strikes, Australian Transport Safety Bureau (ATSB) statistics for the period 2001—2013 indicate only five identified bird strikes at this airport were associated with Pied Herons, out of 1004 bird strikes involving identified species (ATSB 2012, 2014). The flight path to the new roost site at Crocodylus Park docs not cross the airport approach paths, so potential for bird strikes does not currently exist. However, the ability of Pied Herons to change roost sites is demonstrated here, and if in future the birds abandon Crocodylus Park in favour of new harbour roosts then careful monitoring may be required. Roost sites further west than Catalina Island in Darwin Harbour, e.g. the mangroves around Reichardt or Bleesers creeks, may present a more direct threat to aircraft. Roost site fidelity Observations through the 2012-2013 study suggest that the roost site fidelity towards Catalina Island may have broken down during that time. Increased activity in the East Arm of Daman Harbour associated with Inpcx, including substantial night-time activity including bnght lights, may have caused the Pied Herons to change roost sites Alternative nearby roost sites over 2013-2014 include the settlement ponds at East Arm (Amanda Lilleyman pers. comm.) while in some evening surveys Pied Herons flew r er south-west to South Shell Island, or beyond to at least as far as the mangroves ging Wickham Point. These longer commutes between feeding and roosting sites added around 1-5 km to the twice-daily flight of Pied Herons. It t f lr; u n p . obscrvations ,n 2015 show that Picd Hcr ° ns havc aband - cd ^ ***** most s' es, and now roost quite close to the feeding sites at Crocodylus Park. The roost J f ' thc arufic ' al tree-lined river habitat at Crocodylus Park, does not appear to be a limiting factor for Pied Herons, as they only roosted on a fraction of all Liable Habitat use by Pied Herons Northern Territory Naturalist (2016) 27 67 trees. Simon Ferguson, Zoo Supervisor living on-site at Crocodylus Park, noted that the roosting of birds became apparent after completion of the artificial river and also that there had been many Pied Herons roosting at this spot continuously from around May 2014 to the present (September 2015). My single count at the end of September 2015 of about 3300 birds is broadly consistent with the September and October 2013 counts of about 2100 and 3900, indicating that this is a complete count of local roosting Pied Herons, and that the overall population of Pied Herons feeding and roosting in Darwin has not changed dramatically with the change in roost site. It appears that there is a range of roosting options, so the abandonment of Catalina Island, possibly due to disturbance associated with Inpex, docs not appear to be a critical disturbance to the urban population of Pied Herons. Acknowledgements The data presented in this paper have been collected through the efforts of several people, including Matthew Rawsthorne, Kim Rawsthorne, Gavin and Megan O’Brien and Jon Clark, all of whom are heartily thanked. I thank Grahamc Webb for allowing access to Crocodylus Park to observe roosting there, and Simon Ferguson for explaining the recent history of the roost at Crocodylus Park and providing feedback on a draft of this paper. References A'l'SB (2012) Australian aviation wildlife strike statistics. Bird and animal strikes 2002 to 2011. Australian Transport Safety Bureau, Canberra. A'l’SB (2014) Australian aviation wildlife strike statistics 2004 to 2013. Australian Transport Safety Bureau, Canberra. Beehler B.M., Pratt T.K. and Zimmerman D.A. (1986) Birds of New Guinea. Princeton University Press, Princeton, New jersey. Chatto R. (2000) Waterbird Breeding Colonies in the Top End of the Northern Territory. Technical report 6912000. Parks and Wildlife Commission of the Northern Territory, Palmerston. Coates B.J. (1985) The Birds of Papua New Guinea. Dove Publications, Alderly. Garnett S.T. (1985) Heronries of the Mitchell Riv er Delta. Sunbird 15,1-4. Garnett S.T. and Bredl R. (1985) Birds in the vicinity of Edward River Settlement. Parr 1. Introduction, Methods, Study Area, List of Non-passerines. Sunbird 15, 6-23. Marchant S. and Higgins P.J. (eds) (1990) / landbook of Australian. New Zealand and Antarctic Birds (I L4NZAB). Volume I: Rali/es to Ducks. Oxford University Press, Melbourne. McCric N. and Noskc R. (2015) Birds of the Darwin Region. CSIRO Publishing, Clayton South. McKilligan N. (2005) Herons. Egrets and Bitterns. CSIRO Publishing, Collingwood. Morton S.R., Brennan K.G. and Armstrong M.D. (1993) Distribution and abundance of herons, egrets, ibises and spoonbills in the Alligator Rivers Region, Northern Territory. Wildlife Research 20, 23-43. Recher H.F., Holmes R.T., Davis Jr. W.E. and Morton S. (1983) Foraging behavior of Australian Herons. Colonial Watcrbirds 6, 1—10. Robertson B.A. and Hutto R.L. (2006) A framework for understanding ecological traps and an evaluation of existing evidence. Ecology 87, 1075-1085. 68 Northern Territory Naturalist (2016) 27: 68-77 Research Article Evidence of rock kangaroo seed dispersal via faecal seed storage in a tropical monsoon community Christopher T. Martine 1 , Alexandra J. Boni 1 , Elizabeth A. Capaldi 1 , Gemma E. Lionheart 12 and Ingrid E. Jordon-Thaden 13 1 Department of Biological Science, Bucknell University, Lewisburg, Pennsylvania, USA Email: christopher.mar tinc@bucknell.etln Department of Biology, University of Mississippi, University, Mississippi, USA 3 University and Jepson Herbaria, University of California, Berkeley, California, USA Abstract While some of the plant species of the ‘Sandstone Country’ along the escarpment of western Arnhem Land produce fleshy fruits and appear to rely on biotic methods of seed dispersal, little is known about the methods by which this is achieved — and few potential dispersers co-occur in the sandstone outcrop communities. For the present study, scat collections were made on outcrops in the northeastern area of Kakadu National Park with the hope of uncovering relationships between local frugivores and fruit-producers, and providing evidence for seasonal storage of mammal-dispersed seeds in scat pnor to germination. The goals of the present project were to collect and identify sandstone community macropod scat, determine the identity of seeds present in the scat, and provide support for the role of browser/grazer macropods as effective seed dispersers via faecal seed storage in an otherwise disperser-poor local fauna. Scat containing seeds was identified as belonging to the Black Wallaroo {Macropus hern a reins), a rare and locally-endemic macropod considered an intermediate browser/ grazer. These seeds were successfully germinated and the seedlings identified using molecular phylogenetic techniques as Gardenia fucata (Rubtaceac), an endemic rock- specialist species - thus establishing the first confirmation of effective seed dispersal by a rock kangaroo’ in this region and the first identification of a seed disperser for this uncommon Garden,a species. The results provide support for the role of browser/grazer macropods as occasional effective seed dispersers of rock-specialist plant species in the northern monsoon tropics of Australia via faecal seed storage. Introduction Effective seed dispersal is defined not only by the movement of seeds, but also by the °/ T indlVldUalS f ° Ur>Wing tHiU m ° VemCnt Schupp et aL - )• fleshy-fruited plant species relying on endozoochory, identifying effective spcrsal roquets knowledge of whether a given animal ingests the seeds, whether those Seed dispersal by macropods Northern Territory Naturalist (2016) 27 69 Production of fleshy fruits is relatively common among the woody rock-specialist plants occurring in the monsoonal and fire-prone ‘Sandstone Country’ along the escarpment of western Arnhem Land (Northern Territory, Australia). While numerous species appear to rely on endozoochorus seed dispersal, little is known about the methods by which this is achieved in this habitat - and few potential dispersers co-occur with the fruit-producing plants found growing there (Menkhorst & Knight 2011). Rock-dwelling macropods are fairly common in the northern monsoon tropics of Australia and reflect an unusually high regional species diversity there (Telfer & Bowman 2006). The Nabarlek ( Petrogale concinna ), Short-eared Rock Wallaby ( Petrogak brachyotis). Common Wallaroo {Macropus robustus) and Black Wallaroo {Macropus beruardus) (Fig. 1) all occur in sympatry (Menkhorst & Knight 2011). Because these northern ‘rock kangaroo’ taxa are largely nocturnal and generally shy, empirical knowledge of their behaviour, diet and distribution is limited (Richardson 2012). In a 1998 paper, Telfer et at. compiled Aboriginal knowledge of rock kangaroo feeding habits that included numerous anecdotal accounts of opportunistic macropod frugivory - even though the dominant understanding of foraging behavior has defined these animals as intermediate browsers/grazers (Tuft et al. 2011). A later study (Telfer and Bowman 2006) using scat contents to explore niche separation among northern macropods found that at least tine of the species. Macropus bernardus, may be able to inhabit the most rugged and severe habitat because of its ability to utilise leaves, fruits and seeds from a range of rock-specialist plants during dry seasons. In conjunction with this work, Telfer et al (2006) designed a scat identification key to determine current Fig. 1. Black Wallaroo {Macropus bernardus). (Stephen Zozaya, by permission) 70 Northern Territory Naturalist (2016) 27 Martine et al. distribution patterns for macropod species in the Top End region of the Northern Territory; a study that set the stage for a follow-up analysis correlating these distributions with habitat characteristics (Telfer et al. 2008). Coupling the work of Telfer and colleagues with field observations, Martine & Anderson (2007) postulated that rock-dwelling macropods play an important role in short-distance seed dispersal of sandstone endemic plants through a three-step process consisting of fruit ingestion, “faecal seed storage,” and “seasonal redispersal” via wet season rains. However, the authors did not test experimentally whether particular animal taxa actually functioned in this role for specific plant taxa. The goals of the present project were to: 1. Collect and identify sandstone community' macropod scat, examine it for seeds and, if present, germinate them; and 2. Identify the resultant plant(s) species using morphology and, if needed, molecular tools. Materials and Methods In May 2013, 80 scat pellets were collected from upper elevation scat piles on sandstone outcrops in the vicinity of Merl Campground and Cahills Crossing, East Alligator region. Kakadu National Park (Northern Territory, Australia). Two investigators gathered scats using a haphazard sampling scheme during which all intact scats encountered were collected during a single day over a period of roughly eight hours. At the time of collection, the scats were run through the identification key published by Telfer et al. (1996). Although the scats were relatively dry at collection, they were allowed to further air dry' before being packed in paper coin envelopes and shipped to Bucknell University for future work. In July 2013, the scats were dissected and searched for the presence of seeds. Seeds were removed and directly sown into a soil tray without pre-treatment in order to ensure that any germination success might be based on natural processes alone (notably the passage through an animal’s gut and, perhaps, time passed in the scat). Because seedling recruitment on rock outcrops appears to be highest in cracks and fissures where scats/ seeds arc covered with accumulated detritus (CTM, pers. obs.), experimental seeds were shallowly planted below the soil surface. The tray was then placed in a growth chamber witri temperature, light, and humidity- settings previously found to be successful for two w “ endcmic to thc arca where the scats were collected (as per Lionhcart zU14). The soil was kept moist at all times. Once seedlings were apparent, they were transplanted into pots and moved to the uckncll University research greenhouse, where a temperature and light regime matching present conditions in Kakadu was already in place. Following establishment eaf material was removed and dried on silica for future DN A work. Seed dispersal by macropods Northern Terri tor)' Naturalist (2016) 27 71 In April 2014, CTM returned to Kakadu, compared leaves of the germinated greenhouse plants to plants on die site where the scat was collected and, finding a potential match, collected voucher material. This material was examined and identified by staff at the Northern Territory Herbarium in Palmerston, then accessioned at the Manning Herbarium at Bucknell University for later use in DNA extractions. Dried leaf material was extracted using a modified CTAB protocol (Doyle & Doyle 1987). Leaf tissue was pulverised using a GenoGrinder (SPEX Sample Prep) and steel beads in 2 ml microfugc tubes. The CTAB plant solution was incubated for 30 min at 37°C, centrifuged, and moved to a clean tube. The aqueous solution was extracted with chloroform: isoamyl alcohol (24:1) twice. The DNA was precipitated with ice cold 100% isopropanol for 20 min at -20°C, then pelleted for 20 min at 4°C in the centrifuge at 7000 rpm. The DNA pellets were cleaned with two consecutive washes of ethanol, 75% and 95%, respectively, and re-suspended in 100 ml ddH 2 0. The re-suspended pellet was incubated at 37°C with RNasc for one hour, and let to sit overnight at 4°C. The DNA was then frozen at -20°C for storage. The ITS (internal transcribed spacer) gene region was amplified using PCR with the following protocol for a 30 pi reaction volume: 2 min denaturing at 95°C; 29 cycles of denaturing for 30 sec at 95°C, annealing for 30 sec at 56.8°C, and elongation for 30 sec at 72°C; with a final extension for 5 min at 72°C. The reaction mixture for the PCR includes 6 pL GoTaq Flexi buffer (Promega, Madison, WI), 2 pi of MgCl, (1 mM), 1 pi each of forward and reverse primers (0.2 mM, Invitrogen, Carlsbad, CA), 0.6 pi dNTP mix (0.2 mM, Promega), and 1.25 units of GoTaq Flexi (Promega). The primers designed for these studies were as used in Jordon-Thaden et til. (2010) for the ITS of the ribosomal DNA sequence. The primers amplified the 1TS2, ITS1, and the 5.8 S rDNA region. This was done with the 1TS-18 forward 5’-GCA TGT TIT CCC AGT CAC GAC GGA AGG AGA AGT CGI AAC AAG G-3’ which includes an Ml 3 extension (the last 19 bases). The reverse ITS-25 primer was 5’-ACT TCA GGA AAG AGC TAT GAC GGG TAA TCC CGC CTG ACC TGG-3’ which also includes an Ml 3 extension (the first 21 bases). For Sanger sequencing of the ITS region, the Ml 3 extension primer alone was used for the PCR products that had been generated with the Ml3 extension attached to the ITS forward and reverse primers. The M13 extension for forward primer is 5’-GCA TGT TIT CCC AGT CAC G AC-3’ and reverse is 5’-ACT TCA GGA AAC AGC TAT G AC-3’. The PCR products were cleaned with the Promega Wizard cleaning system (Promega), and Sanger sequenced with an ABI sequencer at the Heck Genomics Institute at Pennsylvania State. Sequenced gene regions were processed with Geneious 117 (Biomatters Ltd.) and then compared via BLAST search to confirm the match and the identification of the taxon grown from seeds recovered from the scat. Finding a generic match, multiple ITS accessions of congeners were downloaded, along with an outgroup taxon, and used for phylogenetic comparisons with the two 72 Northern Territory Naturalist (2016) 27 Martine et al “unknown” accessions. Sequences were aligned in Geneious R7 (Biomatters Ltd.) and a maximum likelihood ITS tree was generated with GAR] J (Zwickl et al. 2006), with bootstrap values based on 10,000 reps. Novel sequences generated from the individual collected from scat and the specimen taken from die potential match in die field were deposited in GenBank (see ID numbers in tree figure). Results Use of the Telfer et al. (2006) scat key confirmed that the scat collected on our site was that of the Black Wallaroo, an escarpment-restricted species listed as Near Threatened by the IUCN (Woinarski 2008) largely because of a small total global population limited to a geographic range of about 30,000 km 2 (Telfer & Calaby 2008). Of the 80 scats examined, only one contained seeds, with seven seeds removed from the same sample. These seeds germinated at a rate of 100% without pre-treatment. Plant voucher material collected at the scat collection site was identified by lan Cowie at the Northern territory Herbarium as Gardenia fucata R.Br. ex Benth. (Rubiaceae), a small tree endemic to the sandstone escarpment country (Puttock 1997) in the Top End region of the Northern Territory (Fig. 2). While no previous ITS accession existed in GenBank for G. fucata (or any other Australian Gardenia ), the BLAST search confirmed that the ITS sequences of the greenhouse-grown seedlings (GenBank KP657895) and the wild-collected specimen of G. fucata (GenBank KP657896) were closely allied to other members of the genus Gardenia. Using Gardenia ITS sequences from GenBank (and Ixora pavetta as the outgroup taxon) for comparison to our greenhouse-grown and wild-collected accessions, phylogenetic analysis provided support for their close alliance (l'ig. 4). With this evidence, we can conclude that the seed found in the scat was most likely Gardenia fucata, a species occurring in fairly low' abundance in the site where the scats w'ere collected. Discussion While previous scat analyses (Telfer & Bowman 2006) and compilation of local Indigenous knowledge (Telfer et al. 1998) suggested that Black Wallaroos occasionally eat fruits and seeds, no empirical evidence for then role as effective seed dispersers has been previously established. Our results show' that Black Wallaroos in the East Alligator region of Kakadu National Park occasionally ingest the fruits of Gardenia fucata and that those seeds are not only passed intact in scat but are also able to germinate at a high rate (100%; n=7). In Fig. 2. Developing fruit of Gardenia fucata growing on sandstone outcrop near Cahills Crossing, East Alligator region. Kakadu National Park (CTM 4001). (Chris Martine) Seed dispersal by macropods Northern Territory Naturalist (2016) 27 73 the sandstone habitats where this rock-specialist plant species occurs, few other potential seed dispersers are present - meaning that even if Black Wallaroos (and potentially other rock macropods) are not frequent frugivores, bouts of frugivory and seed dispersal may still play an important role in recruitment of new individuals of Gardenia fucata. The fruits of G. fucata begin ripening in July (peak dry season) and are not especially fleshy, the mesocarp instead having a fibrous nature (Puttock 1997). Local Indigenous knowledge compiled by Telfcr & Garde (2006) includes accounts of M. bernardus (and two other rock kangaroo species) consuming the leaves of G. fucata, but not the fruits. Outcrop interactions The sandstone outcrops of the escarpment country offer a unique suite of underlying opportunities and challenges to their resident flora and fauna. Because outcrops typically retain water, they offer a refuge for plant communities requiring more available resources than the surrounding eucalypt savannah communities (Brock 2001) and thus may support a richer suite of invertebrate and vertebrate animals. Outcrops support communities that are distinct from the surrounding landscape - including numerous endemic species (Brennan 1986; Brock 2001). At the same time, these outcrop communities, while similar to the nearby plateau country, also remain distinct from it by virtue of their smaller size and island-like isolation. In the most pronounced cases they serve as refugia for drought-intolerant species that, formerly more widespread, were forced to retreat to these islands of tolerable habitat during the Pleistocene - the best example perhaps being Al/osyncarpia ternata, a tree common in previously widespread rainforests that is now found only in refugial monsoonal gorges along the Arnhem Plateau escarpment (Russell-Smith 2009). These restricted rock outcrops may have also served (and continue to serve) as refugia for Macropus bernardus. Telfer & Bowman (2006) suggest that M. bernardus, the species with smallest range (Fig. 3) of six Macropus species in Australia (Woinarski 2008; Coulson and nidridge 2010), was forced to retreat to these habitats during glacial periods of the Pleistocene and subsequently adapted to exploit rock-specialist plants (many of them chemically-defended). All rock kangaroos are considered refuge-dependent, with disjunct occurrences across broad geographic regions reflecting the scattered distribution of compatible habitat (Tuft et aL 2011); and this is strongly pronounced in M. bernardus. An analysis of environmental correlates found that M. bernardus distributions, more so than the patterns for three sympatric rock-dwelling macropods, are especially linked to the presence of rugged, rocky terrain (Telfer et al 2008). For plant communities on outcrops, these habitats operate like terrestrial islands, with isolated populations for which gene exchange is challenged through the limited movement of pollinators and seed dispersers (Martine & Anderson 2007; Roche et al 2014). While little has been done on the effects of outcrop-to-outcrop distance on gene exchange, the relationship between two sandstone specialist sister species. Solarium asymine triply Hum and S. sejunctnm, shows that the taxa remain distinct (Brennan et al. 2006; Martine et al. 2006) even though they can hybridise and are separated by just 10 km at the nearest point in their combined distribution (Gilman et al. 2014). Previous inferences (Anderson 74 Northern Territory Naturalist (2016) 27 Martine et al. & Symon 1988; Martine & Anderson 2007; Martine & Capaldi 2013), have noted that, although long-distance pollinators arc rare visitors to flowers on outcrops, most of the services arc provided by small-bodied bees flying over short distances. Although we can assume that gene migration is also limited by seed dispersal challenges (Symon 1979; Martine & Anderson 2007), little has been done to show it. Martine & Anderson (2007) theorised that rock wallabies were key players in the movement of outcrop seeds, but recognised that this was likely to occur within outcrops rather than between them. Because these macropods are poorly adapted for movement across flat ground (a consequence of being exceptionally adapted for climbing around rocks) (Telfcr & Bowman 2006), they are most likely to facilitate short distance dispersal within outcrops. This might be accomplished through nightly dry season foraging bouts where seeds arc moved into ‘faecal seed storage’ in/around daytime roosting spots (Telfer & Griffiths 2006), followed by ‘seasonal redispersal’ via wet season downpours — an inference supported by the frequent occurrence of young plants in steep washes, ravines and deep cracks (CTM, pers. obs.). The present study shows that rock-dwelling macropods may occasionally function as effective seed dispersers for at least one sandstone outcrop endemic and, by nature of the locations and manner in which seeds were collected, appear to provide faecal seed storage as per Martine & Anderson (2007). Occasional seed and fruit opportunism on the part of the macropods (sec Jarman 1994; Telfcr & Bowman 2006) may thus be as important to the plants as to the animals, particularly during seasons when the availability of grasses and foliage is reduced. Whether this interaction then leads to seasonal redispersal during the monsoon season is left to be determined, but knowing that seeds stored in faeces on upper elevation rocks can germinate without any additional treatment aside from the presence of a moist substrate allows us to assume that seedling recruitment can follow redistribution ot seeds by water and gravity. Telfcr et al. (2006) found that although scats of the Short¬ tailed Rock Wallaby may persist in nature for over two years, most scats are lost within six months, ostensibly being degraded and washed away by wet season rains. Faecal seeds redispcrsed during these rains might then reap the benefits of increased access to nutrients and carbon supplied by the faeces (Kobayashi et al 2011). I Fig. 3. Map of the Top End region of the Northern Territory showing approximate range of Macropus bernardus (red) based on Menkorst & Knight (2011) and locations (dots) of Gardenia fucata collections accessioned at the Northern Territory Herbarium (from 1. Cowie). (Map drawn with ESRJ ArcGIS) Seed dispersal by macropods Northern Territory Naturalist (2016) 27 75 Gardama fucata Gardenia lasminowes Given the low numbers of seeds — recovered in this study, we cannot discount the potential for post-dispersal seed predation to render our findings moot. Ants and rodents are known to predate seeds released in scat (e.g. Hulme 1998; Manzano et al 2010; Pcco et at 2014); and both are in abundance in the study area. However, seeds moved but not consumed during bouts of post-dispersal predation (see Azacarate & Manzano 2011) may benefit from redispersal similar to that hypothesised above. There is growing evidence that small- and medium-sized mammals throughout the northern monsoon tropics of Australia have recendy experienced and continue to face major declines, even in Kakadu National Park and other protected areas considered to be ecologically intact (Woinarski et al 2001; Ziembicki et al 2012; Woinarski & Fisher 2013). Further work is needed to help understand and protect the links between these mammals and the plants that may depend on them, at least in part, for their long-term survival. Descriptive studies like this one provide individually small steps that should prove cumulatively integral to conservation efforts, particularly for organisms whose natural histories are as poorly understood as those highlighted here. Conclusions Fig. 4. Maximum likelihood ITS tree of the plant grown from scat collections, Gardenia sp. GAR01, in relation to other Gardenia species and selected outgroup taxon, Ixora pavetta. Bootstrap values based on 10,000 reps. GenBank: Scat-grown-GAllOl: KP657895; Gardenia fucata CTM4001: KP657896; G. hansemannir. FM204691; G.jasminoidey. KC533838; G. thunbergia: AJ224833; and Ixora pavetta-. J X856466. Treebank ID forthcoming. This study establishes the potential role of rock macropods as dispersers of rock- specialist plants in the monsoon tropics of Australia. Scat containing seeds was identified as belonging to the Black Wallaroo, a rare and locally-endemic macropod considered an intermediate browser/grazer. Seeds were successfully germinated and seedlings were identified using molecular techniques as Gardenia fucata , an endemic rock- specialist species - thus establishing the first confirmation of effective seed dispersal by a rock kangaroo in this region and the first identification of a seed disperser for this uncommon Gardenia species. The results provide support for the role of browser/grazer macropods as occasional effective seed dispersers of rock-specialist plant species via faecal seed storage. Herbarium voucher information: Gardenia fucata R. Br. ex Benth. C.T. Martine 4001. 9 April 2014. Outcrops near Merl Campground, Kakadu National Park, Northern Territory, Australia. 76 Northern Terri tor)' Naturalist (2016) 27 Martine et al. Acknowledgements Anne O’Dca, Kakadu National Park, provided CTM assistance with obtaining permits and accommodation. The staff of the Northern Territory Herbarium provided assistance with collections, shipments and cups of tea; and we are especially indebted to Deborah Bisa. Field assistance: Rachel Martine, Iscc Martine, Jackson Martine and C. Erin Sullivan. Additional lab and greenhouse support: Wanda Boop, Tara Caton, Alex Ehrenman, Emma Frawley, Lacey Gavala, Ian Gilman, Dan Hayes, Sebastian Martinez, Morgan Roche and Mark Spiro. Funding provided by the David Burpee Endowment and the David T Scaddcn Faculty Scholar Award (to CTM and EC) at Buckncll University with additional support to AJB and GL trom the Botanical Society of America Undergraduate Student Research Award and the Wayne E. Manning Internship Fund at Buckncll University. References Anderson G.J. and Symon D.E. (1988) Insect foragers on Solatium flowers in Australia. Annals of the Missouri Botanical Garden 75, 842-852. Azcarate EM., and Manzano P. (2011) A fable on voracious and gourmet ants: Ant-seed interactions from predation to dispersal. In: Predation in the Hymenoptera: An evolutionary perspective, (ed. Polidori C). Transworld Research Network, Kerala. Brennan K. (1986) IP itdflomrs of Kakadu: A guide to the witdfimvcrs of Kakadu National Park and the Top '■ndo/ the Northern Territory. Published by the author. Brennan K„ Martine C.T. and Symon D.E. (2006) Solatium sejunctunr. A new dioecious species from lWf/OTf U '’' ,N l°-7 hern 1 trrltory ' Tbt Dea £ /e - Rttords of the Northern Territory Museum of Arts and C ° U Qayton, VktorifcIlRO. ^ ^ ° f "' ullMes ' and "‘-kangaroos. Doyle j. J . and Doyle J.I (1987) A rapid DNA isolation procedure for small quantities of fresh leaf tissue. P/ylochemcal Bulletin 19, 11-15. ' G ^e n nrodu^ tin i C io < r I J° rdon ? ,ad ] en 1E - (2014) Vague species boundaries exposed: ^olny20H^tmJul!ZT ^ ™ ^ in AuStraUan Hul me P.E. (1998) Post dispersal seed predation: consequences for plant demography and c\o uuon. / erspectim in Plant Geology, Iivo/ution and Systematic.! 1, 32-46. 5 J arman Pj^(1994) The eating of seedheads by species of Macropodidae. Australian Mammalogy R ° C thr^e breeding systems 'and ^ (201 , 4) Po P ula “™ S-eric comparisons across ID, July 2014. UC SptCltS ln Aui,tralla - Solatium. Abstract. Botany 2014, Boise, W p *“ " ( («-*- w) a fv Undei^raduate Seed dispersal by macropods Northern Territory Naturalist (2016) 27 77 Manzano P., Azcarate F.M, Peco B., and Malo J.E. (2010) Are ecologists blind to small things? The missed stories on non-tropical seed predation on feces. Oikos 119, 1537—1545. Martine C.T., Vanderpool D., Anderson G.J., and Les D.H. (2006) Phylogenetic relationships of andromonoecious and dioecious Australian species of Solatium subgenus Ijptostemonum section Mekmgfmr. Inferences from ITS sequence data. Systematic Botany 31, 410—420. Martine C.T. and Anderson G.J. (2007) Dioccy, pollination, and seed dispersal in Australian spiny Solatium. I 'Ith International Sotanaceae Conference: Acta Horticullurae 745, 269-283. Martine C.T. and Capaldi E.A. (2013) Is inaperturate pollen produced by Australian dioecious Solanum a false reward for pollen foraging bees? Abstract. Botany 2013, New Orleans, LA, July 26-30, 2013. Menkhorst P. and Knight F. (2011) A field guide to the mammals of Australia , 3 ,d edn.. Oxford University Press, Oxford. Peco B., I.affan S.W. and Moles A.T. (2014) Global patterns in post-dispersal seed removal by invertebrates and vertebrates. PI j/S ONE 9, e91256. doi:10.1371/journal.pone.0091256. Puttock C.F. (1997) A revision of Gardenia (Rubiaceae) from northern and north-western Australia. Nuytsia 11,225-262. Richardson K. (2012) Australia’s amazing kangaroos: their conservation, unique ecology and coexistence with humans. CSIRO, Collingwood, Victoria. Russell-Smith )., Lucas D.E., Brock J., and Bowman D.M.J.S. (2009) Allosyncarpia-domu\Mci3 rain forest in monsoonal northern Australia, journal of Vegetation Science DOl: 10.2307/3235734. Schupp 11.W., Jordano P. and Gomez J.M. (2010) Seed dispersal effectiveness revisited: a conceptual review. New Plpilologist 188, 333-353. Symon D.E. (1979) Fruit diversity and dispersal in Solanum in Australia. Journal of the Adelaide Botanic Garden 1, 321-331. Telfer W.R. and Bowman D.M.J.S. (2006) Diet of four rock-dwelling macropods in the Australian monsoon tropics. Austral Ecology 31,817-827. Telfer W.R., Griffiths A.D. and Bowman D.M.J.S. (2006) Scats can reveal the presence and habitat use of cryptic rock-dwelling macropods. Australian Journal of Zoology DOl: 10.1071 /Z005074. Telfer W.R. and Garde M.J. (2006) Indigenous knowledge of rock kangaroo ecology in western Arnhem Land, Australia. I iuman Ecology 34, 379-406. Telfer W.R. and Griffiths A.D. (2006) Dry season use of space, habitats and shelters by the short¬ eared rock-wallaby, Petroga/e hradryotis, in the monsoon tropics. Wildlife Research 33, 207-214. Telfer, W.R. and Calaby J.H. (2008) Black wallaroo. Macropus liernardus. In: The mammals of Australia. 3rd edn. (cds Van Dyck S. and Strahan R). Reed New Holland, Sydney. Telfer Will., Griffiths A.D. and Bowman D.M.J.S. (2008) The habitat requirements of four sympatric rock-dwelling macropods of the Australian monsoon tropics. Austral Ecology 33 1033-1044. l'uft K.D., Crowther M.S. and McArthur C. (2011) Multiple scales of diet selection by brush-tailed rock-wallabies ( Petroga/e penicillata). Australian Mammalogy 33,169-180. Woinarski J.C.Z., Milne D.J. and Wangancen G. (2001) Changes in mammal populations in relatively intact landscapes of Kakadu National Park, Northern Territory, Australia. Austral Ecology 26, 360—370. Woinarski J.C.Z. (2008) Macropus bernardus. In: IUCN 2013. IUCN Red List of Threatened Species. Version 2013.2 (accessed 28 May 2014) Woinarski J.C.Z. and Fisher A. (2013) Threatened terrestrial animals of Kakadu National Park: Which species?; How are they faring? And what needs to be done for them? In: Kakadu National Park lumdscape Symposia Series. Symposium 7: Conservation of threatened species, (eds Winderlich S. and Woinarski J.) Internal Report 623. Supervising Scientist, Darwin. Northern Territory Naturalist (2016) 27: 78-83 Short tSI ote Report of the presence of Hapalotrema synorchis and H. postorchis (Digenea: Spirorchiidae) in marine turtles (Reptilia: Cheloniidae) in Northern Territory waters Diane P. Barton 1 , Phoebe A. Chapman 2 and Rachel A. Groom 3 1 Natural Sciences, Museum & Art Gallery of the Northern Territory, GPO Box 4646, Darwin, NT 0801, Australia Email: di.barton@,nt.gov.au Veterinary-Marine Animal Research Teaching and Investigation Unit, School of Veterinary Science, The University of Queensland, Gatton, QLD 4343, Australia Marine Ecosystems, Flora and Fauna Division, Department of Environment and Natural Resources, PO Box 496, Palmerston, NT 0831, Australia Abstract The spirorchiid digenean Hapalotrema synorchis was recovered from the heart of a juvenile Hawksbill Turtle ( Eretmochelys imbricata) found deceased on a local Darwin beach. The turtle was in poor condition, showing many characteristics associated with spirorchiid infection. A second necropsied turtle showed signs of infection with spirorchiids but adult specimens were not recovered. Examination of specimens held at the Berrimah Veterinary Laboratory found another E. imbricata infected with H. synorchis and a Green Turtle (Chelomamydas) infected with the related H. postorchis. Despite previous reports of infected turdes, this is the first confirmed identificadon of H. synorchis and H. postorchis from Northern Territory waters. Manne turtles arc hosts to spirorchiid digeneans that inhabit the cardiovascular system including the tissues and blood vessels of all the major organs (Glazebrook Hal. logty C hapman eta! 2015; Flint eta! 2015). Spirorchiids have been implicated as causes of standings and mortality in turdes around the world, with infected turtles often showing T/ 20uJ P " and 8CnCral liStk8SnCSS ( Glai!ebr <>ok rial 1989; Stacy Clinical signs of infecdon with spirorchiids include sunken eyes, plastron shrinkage and generalised muscle wastage (Glazebrook rial 1981; Gordon ItaL " 98 ) and mime ses, neurological symptoms (Jacobsen ri al 2006). However, the direct correlation O spirorc 11 infection with disease in turdes is poorly understood (Stacy et a / °010) c n f r to harb J ^ ££1 45% —-* s.”' / lapalotrema in marine turtles Northern Territory Naturalist (2016) 27 79 for necropsy were infected with spirorchiids, which were deemed the primary cause of death in around 42% of cases. Egg granulomas can be found in high numbers in turtles where adult digeneans arc apparendy absent (Gordon et al 1998; Flint etaL 2010). Although reports of cardiovascular digeneans in stranded marine turdes from Northern Territory waters are known (l.impus 2009; Mackouras & Griffiths 2014), no species- level idcndfication has been undertaken. Mackouras & Griffiths (2014) reported on the pathological examination of eleven marine turtles stranded from lune 2012 to )unc 2014 in Northern Territory waters. One of six Green Turtles (< Chelonia mydas) was found to have died as a result of severe cardiovascular digenean infection; two of five I lawksbill Turtles ( Eretmocbetys imbricata) were infected with cardiovascular digeneans, but these were not considered the cause of death. A juvenile E. imbricata , was submitted to the Berrimah Veterinary Laboratories (BVL) for post-mortem assessment after it was found deceased on a local beach (in the suburb of Nightcliff (12.3783°S, 130.8453°E) on 10 July 2014. The turtle weighed 2.5 kg and externally appeared thin and in poor condition. During necropsy it was noted that the abdominal fat was absent except for remnants attached to the plastron (lower shell) and around organs and there was extensive clear fluid present within the abdomen and pericardial sac. Granulomas containing digenean eggs were present in the blood vessels associated with the lungs, liver, heart, intestine and fat. Digeneans were recovered from the heart and placed into a vial of 70% ethanol. A second (female) E. imbricata was submitted to the BVL after it was found deceased entangled in a net at Ixe Point (12.3292°S, 130.8844°E) on 8 |anuary 2015. The turde weighed 3.5 kg and had a curved carapace length of 35 cm. The turtle was in good body condition. No digeneans were recovered from the heart or lungs. A small number of egg granulomas were observed in the blood vessels of the intestinal system. A subsequent search of the BVL parasite collection found two vials of digeneans identified as Hapa/otrema sp. The first came from an Eretmochelys imbricata collected from Bare Sand Island, Fog Bay (12.5369°S, 130.4189°E) in 1997 and the second from a Chelonia mydas (collection data listed as Darwin) in 2003. From each sample, the anterior end (anterior to the ventral sucker) of one specimen was removed, placed into 100% ethanol, and used for genetic analysis. The remaining portion of that specimen was stained with accto-orcein, dehydrated through a graded ethanol series and mounted in Canada Balsam. The 1997 E. imbricata vial contained a number of specimens, so an intact specimen was also mounted. Further specimens were left intact as unmounted specimens. The digeneans recovered from the 2014 E. imbricata were identified as Hapa/otrema synorchis through a combination of morphological and genetic analyses. The specimens recovered from the 1997 E. imbricata were also identified as H. synorchis and the specimens from the Chelonia mydas were identified as H. postorchis through morphology; unfortunately the 80 Northern Territory Naturalist (2016) 27 Barton et al. Fig. 1. \ iapalotrema synorch/s. Lateral view of whole specimen orientated upside-down as is Fig. 2, the convention for digeneans. Abbreviations: OS, oral sucker; VG, vitelline (= yolk) glands; VS, ventral sucker. Scale bar = 1 mm. (Adam Bourke) V T al V ' CW ” f SUined whoIe mount specimen. Abbreviations: (= yolk) Xds VS T Un f “c'TV ’ SUckcf ’ PT ’ P° stcrior tcstts - VG, vitelline 1 y J glands, \ S, ventral sucker. Scale bar = 1 mm. (Adam Bourke) •genetic analysis was unsuccessful on these specimens, possibly because they may have been initially preserved in formalin. b«ed ™ Z” “'IT 1 ,hiS snjdr WCre '““t identified to the gene, H.pahmm. bated „„ die motphologtcal cmena of the gene, (tee Plan 2002). Morphologicall, the Bt(|» Zh ", hil>i ' Cd Z * indicated in Plat,'and ZS H : T «*'* f »™S compact ameno, and pottterio, masses. mcZ*TT T “■ '**«*«•»* using the PCR and segneneng Z“Zd „ 1 ■" T™” " “ l (2 ° ,5) *” A BLAs! search <99% similarity u r “ U l "' B sequence, which indicated that the closest match (99/, similarity) „as H. gmOi. TV spectntens from CM* w „e tdendfied Wdpalotrema in marine turtles Northern Territory Ndturdlist (2016) 27 81 as H. postorcbis due to a similar number and arrangement of the testes (16 counted; 9 posteriorly, 7 anteriorly) and the vitelline glands as described by Dailey et al. (1993). Hapalotrema synorchis was described by Luhmann in 1935, from a Loggerhead Turtle (Caretta caretta) from Tortugas, Florida (Platt & Blair 1998). Subsequent records of H. synorchis occurred from Chelonia mydas in the Gulf of Mexico (Caballero y Caballero 1962) and from E. imbrica/a in Puerto Rico (Fischthal & Acholonu 1976). Takeuti (1942) described H. orientaJis from E. japonica (= E. imbricata) in Japanese waters, which was subsequently synonymised with H. synorchis by Platt & Blair (1998). Platt & Blair (1998) reported C. caretta as a host for H. synorchis at Shark Bay, Western Australia (4 turdes), Heron Island (2) and Mon Repos (1) in central Queensland waters. Chapman et al. (2015) subsequently reported H. synorchis from E. imbricata from Redland Bay and Buddina Beach in south-east Queensland. In the study by Chapman et al. (2015), H. synorchis was found in two of three E. imbricata , but no Chelonia mydas from Queensland (of 22) or Hawaiian (of 10) waters. Stacy (2008) also reported H. synorchis from C. caretta from Florida (originally described as I I. pambanensis, however later amended to H. synorchis when further sequences became available for comparative molecular analysis (see Chapman et al. 2015). Hapalotrema postorchis was described by Rao from Chelonia mydas in the Gulf of Manar, India (Dailey et al. 1993). It has subsequendy only been reported from the same host species in Hawaii (Dailey et a,L 1993; Chapman et at. 2015: one of ten hosts examined) and various locations in south-east (Morcton Bay; Coolum) and central Queensland (Quoin Island, Gladstone Harbour) (Cribb & Gordon 1998: three infections recorded from an unstated total number of hosts examined; Gordon et al 1998: three infections from 96 hosts examined; Chapman et ai 2015: total of three from 22 hosts examined). Reports of infections with Hapalotrema from Australian waters arc confusing as many reports of digeneans infecting turtles were undertaken prior to the review of the genus (see Cribb & Gordon 1998). Glazcbrook et al. (1981) reported a Hapalotrema sp. from a heavily infected Chelonia mydas off Townsville, North Queensland. Although they stated that it was a different species to H. synorchis, Glazcbrook et al. (1989) subsequently listed it as H. synorchis in their literature records of cardiovascular digeneans recovered from sea turtles. However, H. synorchis has never been reported from another Chelonia mydas in Australia, despite a large number of this turtle species being examined (see Chapman et aI. 2015). Morphologically, H. synorchis is very similar to H. pambanensis (a synonym of H. mehrai), so it is more likely, without genetic confirmation, that the original record from Glazebrook et al. (1981) belonged to this species. Four species of marine turtles are found in the waters of the Northern Territory (Mackouras & Griffiths 2014). During the period 2012-2014, a total of 60 marine turtles were reported as stranded in the Northern Territory; the most common species was the Green Turtle ( Chelonia mydas-, 26) followed by the Hawksbill (E. imbricata-, 16) (Mackouras & Griffiths 2014). The vast coastline of the Northern Territory and the array of large 82 Northern Territory Naturalist (2016) 27 Barton et al. marine predators (saltwater crocodiles and sharks) unfortunately limits the access to fresh carcasses required for necropsy for a systematic survey of diseases of marine turdes. During the 2012—2014 reporting period, 11 turtles were necropsied at the BVL. Of these, one (of six) Chelonia mydas was determined to have died from a severe cardiovascular digenean infection and one (of five) E. imbricata contained a heavy cardiovascular digenean infection, but this was not the suspected cause of death (Mackouras & Griffiths 2014). A further E. imbricata contained possible cardiovascular digenean egg granulomas, but no adult digcncans were observed (Mackouras & Griffiths 2014). Unfortunately, no digencans specimens were available from these necropsies for examination in this study. A study into stranded Green Turdes from Gladstone, central Queensland, found that nine (of twelve) turdes were infected with spirorchiid digeneans at a level that could have contributed to their death (Hint et al 2015). All turtles, however, were infected with a variety of spirorchiids that belonged to five genera, including Hapalotrema. Hapalotrema synorebis is a widespread parasite that has been positively identified from three species of marine turtles in waters of Queensland and Western Australia, Japan, the Gulf of Mexico, Horida and Puerto Rico. Hapalotrema postorchis has only been reported in Chelonia mydas , with a distribution in w r atcrs of India, Queensland and Hawaii. This is the first confirmed report of the presence of H. synorchis and H. postorchis in waters of the Northern Territory. I he following voucher specimens have been deposited in the collections of the Museum and Art Gallery of the Northern Territory: II. synorchis from E. imbricata (2014) (D1532); / /. synorchis from / imbricata (1997) (D1545); H. postorchis from Chelonia mydas (D1546). The Genbank accession number for the //. synorchis is KT361641. Acknowledgments The authors are indebted to the staff at the BVL for access to specimens from turtle necropsies. The authors are also grateful to the Northern Territory Government Berrimah harm Library staff for their speedy acquisition of reference material. References " ' , * A b :. cr ° E ; £962) Trcmatodos de las tortugas dc Mexico. X. Prcsencia de Orchidasma amphwnbis (Braun, 1899) Looss, 1900 en una tortuga marina. Che/one mydas del las costas del estado dc 1 ampaulipas, Mexico. Anates de la Instilula de Bio/ogia, Mexico 33, 47-55. Chapman P A., Crtbb LH Blair D., Traub R.J., Kyaw-Tanner M.T., Flint M. and Mills PC. (2015) Molecular analysis of the genera Hapalotrema Looss, 1899 and Uaredius Price, 1934 (Digcnea: Wi,h ““ “ *• »“* * 8™* Cdh (cl^:: d ^° rd : n ( ’" 8) Spirorchidae) in the green turde (Cbetoma mydas) in Australia, journal of Parasitology 84, 375-378. Hapalotrema in marine turtles Northern Territory Naturalist (2016) 27 83 Dailey M.D., Fast M.L. and Balazs G.H. (1993) Hapalotrema dorsopora sp. n. (Trematoda: Spirorchidae) from the heart of the green turde {Chelonia mydas) with a redescription of Hapalotrema postorchis. Journal of the Helminthological Society of Washington 60, 5—9. Fischthal J.H. and Acholonu A.D. (1976) Some digenetic trematodes from the Adantic hawksbill turtle, Eretmochefys imhricata imhricata (L.), from Puerto Rico. Proceedings of the Helminthological Society of Washington 43,174-185. Flint M., Eden P.A., Limpus C.J., Owen H., Gaus C. and Mills P. C. (2015). Clinical and pathological findings in green turtles ( Chelonia mydas) from Gladstone, Queensland: Investigations of a stranding epidemic. EcoHealth 12, 298-309. Flint M., Patterson-Kane J.C., I.impus C. and Mills P.C. (2010). Health surveillance of stranded green turdes in southern Queensland, Australia (2006-2009): an epidemiological analysis of causes of disease and mortality. Ecohealth 7, 135—145. Glazebrook J.S., Campbell R.S.F. and Blair D. (1981) Pathological changes associated with cardiovascular trematodes (Digenca: Spirorchiidae) in a green sea turtle Chelonia mydas (L). Journal of Comparative Pathology 91, 361—368. Glazebrook J.S., Campbell R.S.F. and Blair D. (1989) Studies on cardiovascular fluke (Digenea: Spirorchiidae) infections in sea turtles from the Great Barrier Reef, Queensland, Australia. Journal of Comparative Pathology 101,231-250. Gordon A.N., Kelly W.R. and Cribb T.H. (1998) Lesions caused by cardiovascular flukes (Digenea: Spirorchidae) in stranded green turdes {Chelonia mydas). Veterinary Pathology 35, 21-30. Jacobson E.R., Homer B.L., Stacy B.A., Greiner E.C. et at. (2006) Neurological disease in wild loggerhead sea turdes Caretta caretta. Diseases of Aquatic Organisms 70, 139-154 1 .lrnpus C.J. (2009) A biological review of Australian marine turtle species. 3.1 lawkshi/l turtle, Eretmochelys imhricata (Linnaeus). The State of Queensland Environmental Protection Agency. Mackouras K. and Griffiths A.D. (2014) Northern Territory Marine Megafauna stranding!: 2012 to 2014. Report to Department of Land Resource Management, Darwin. Platt T.R. (2002) Family Spirorchiidae Stunkard, 1921. In: Keys to the Trematoda, Vol. 1 (eds Gibson D. I., Jones A. and Bray R.A.), pp. 453-467. CAB International. Platt T.R. and Blair D. (1998) Redescription of Hapalotrema mistroides (Monticelli, 1896) and I iapalotrema sytwrchis Luhman, 1935 (Digenea: Spororchidae), with comments on other species in the genus. Journal of Parasitology 84, 594-600. Stacy B.A. (2008) Spirorchiid trematodes of sea turtles in Florida: Associated disease, diversity, and life cycle studies. Unpublished PhD Thesis, University of Florida, Gainsville, Florida. Stacy B.A., Foley A.M., Greiner E., Herbst L.H., Bolten A., Klein P., Manire C.A. and Jacobsen E. R. (2010) Spirorchiidiasis in stranded loggerhead Caretta caretta and green turtles Chelonia mydas in Florida (USA): host pathology and significance. Diseases of Aquatic Organisms 89, 237-259. Takeuti E. (1942) New blood-flukes of the family Spirorchidae from Japanese fresh water tortoise and marine turtles . Japanese Journal of Medical Science, bacteriology and Parasitology 12, 161-174. Northern Territory' Naturalist (2016) 27: 84—96 Research Article Coral communities in extreme environmental conditions in the Northern Territory, Australia Lawrance W. Ferns Biodiversity Division, Department of Environment, I .and, Water and Planning, 8 Nicholson Street, East Melbourne, Victoria 3002, Australia Email: lawrance.fcrns@delwp.vic.gov.au Abstract An extensive intertidal reef flat in the macro-tidal marine waters of the Northern Territory was chosen to investigate species composition and zonation persisting under extreme environmental conditions. Thirty-six visual belt transects were used to quantify scleractininan corals, benthic algae and other sessile invertebrates which varied in vertical and horizontal space. Thirty-four coral species were identified. Most species were represented by the family Merulinidae, with lifeform characteristics typical of species specialised in environmental tolerance to high sedimentation, turbidity and temperature (i.e. massive, sub-massive and encrusting growth forms with convex and steep sided morphologies, thick skeletal tissue and large polyps). Whilst the combination of environmental and ecological characteristics of this reef flat community can be viewed as distinctive to the Darwin region, a number of similarities can be compared to reef communities reported in extreme environments of the Arabian Gulf, Red Sea and other regions of tropical northern Australia. Introduction (.oral communities which presently persist in extreme environmental conditions arc of contemporary interest towards understanding species resilience and potential for adaptation to climate change (Hughes et al 2003; Bauman et al 2011, 2013a, 2013b; Dandan et al 2015). Predictions for climate change stressors for coral communities include warmer sea temperature and changes in extreme episodic events such as heavy rainfall, storms and possible sediment and nutrient debouching from rivers and run-off (( tilmour et al. 2006). Such environmental perturbations are known to be associated with mass coral bleaching and sudden die off (Glynn 1993; Depczynsld et al. 2013). Coral species of the coastal waters in the vicinity of Darwin Harbour, Northern Territory (Wolstenholme et al. 1997), the Kimberley coast, northwestern Australia (Dandan et al. 2015; Schoepf et al 2015) and the Arabian Gulf (Sheppard & Sheppard 1991; Coles 1997; Coles 2003; Riegl 1999; Bauman dal. 2011; Riegl & Purkis 2012) survive in harsh conditions and offer insight to species with high climatic tolerances and adaptation. Climatic trends in this region of Australia are changing. Since 1950, the Northern Territory average rainfall has risen 35.7 mm per decade during November-April and Coral communities in extreme environments Northern Terri tor} 1 Naturalist (2016) 27 85 fallen 0.4 mm per decade during May-October. From 1910 to 2003, the intensity of heavy daily rainfall gradually rose by 10% (Hennessy et al. 2004). Water temperatures at the shoreline may reach over 36°C at high tide, with tide pools and standing water bodies reaching over 43 °C at low tide which is similar to the high temperature fluctuations in the Arabian Gulf, and well above temperature ranges corals traditionally considered limiting to coral survival (Coles 1997). Sea surface temperatures north of Australia have been at record-breaking highs. In 2010, temperatures north of Australia broke previous records by large margins and were also above average during the 2011-2012 La Nina event (Australian bureau of Meteorology 2013). An extensive intertidal reef flat community off the north-eastern shoreline of Darwin Harbour was chosen to investigate coral zonation on a macro-tidal shoreline. The marine waters of Darwin Harbour are subjected to daily and seasonal fluctuations in sea surface temperature, light availability and extreme levels of sedimentation and turbidity which provides insight to northern Australian coral species that have already adapted to extreme climatic conditions. Methods Study site Nightcliff Reef (Fig. 1) was surveyed at low spring tide between September and October 1994. Darwin Harbour is a ria coast formed by the post-glacial marine flooding of a dissected plateau. Subsequent sedimentary infill has resulted in the formation of numerous embayments, islands and extensive mangrove-vegetated tidal flats (Semeniuk 1985). The dominant lithological type is ferricrete laterite with phyllite/siltstonc and this is reflected by the presence of conspicuous medium to coarse grained lateritic pebbles of the upper beach sediments. The coral community at Nightcliff exists as a veneer reef by colonising hard substrates of rock and consolidated materials without accreting substantial calcium carbonate substrata (Hooper 1987; Mitchie 1987). The community extends seaward to a distance of approximately 500 m from the shoreline at low spring tide. Tides in Darwin Harbour exhibit semidiurnal inequality with a spring tidal range is in the vicinity of 0.1-7.8 m. Tidal currents are very strong ranging from 0.25-1.4 m/s and recording as high as 2 m/s (Semeniuk 1985). The concentration of rivers, streams and creeks with accompanying discharge, and the strong tidal movement of the system, account for excessive sediment mixing and high turbidity (Michie 1987). Nightcliff Reef is fully cmersed for 2-3 hours during low spring tides that are less than 1 m. Low spring tides occur between 1100-1600 hrs in the period September to March (wet season) and between 2300-0400 hrs in the period April to August (dry season). Stratification A topographic contour map of the reef was generated from random spot heights measured using a digital theodolite. All heights were referenced from a survey datum 86 Northern Territory Naturalist (2016) 27 Ferns point adjacent to Nightcliff Pier, and were converted to metres above the lowest astronomical tide (i.e. 0 m). The contour map was used to stratify the reef into three main vertical zones: (1) the upper reef flat which occupied a vertical height between 1.4—1.8 m; (2) the middle reef flat between 0.9-1.4 m; and (3) the lower reef flat between 0.3—0.9 m. Sampling Sampling was conducted using four replicate 1 m x 20 m contiguous belt transects that were haphazardly placed in each of the upper, middle, and lower reef flat vertical zones 1 he sampling was repeated in three localities (north, central and south locations) to account for any variations in substrata micro-topography, sediment deposition and taxa composition that may occur horizontally across the reef relative to the shoreline. In total, 36 belt transects were surveyed. Contiguous belt transect was chosen as the sampling unit to improve the recording of small and less abundant species, and representation of microhabitats (e.g. large or small coral colonies, sand patches, tide pools) in any given zone and locality (Chiappone & Sullivan 1991; Sullivan & Chiappone 1992). Species abundances of all sessile taxa were recorded by visual estimates of percentage cover and numbers of individuals for each 1 m~ quadrat. Lifeform attributes of coral colonies were recorded using categories of English et al (1994). When a particular species was encountered that could not be identified in the field, a sample of the species was collected for identification in the laboratory and also compared to collection specimens held at the Northern Territory Museum. Original taxonomic identifications for scleractinian corals followed that of Veron and Pichon (1976, 1980, 1982); Veron et ti! (19’ 7), Veron and Wallace (1985) and Veron (1986). Algal identifications followed that of Jaasund (1976); Cribb (1983); Cnbb and Cribb (1985); Lawson and John (1987) and Pncc and Scott (1992). Updates to recent taxonomic revisions follow Wynne (2011); Budd et aL (2012); Guiry and Guiry (2013); Huang et al. (2014) and WoRMS Editorial Board (2015). All data presented in the text and figures are the arithmetic mean. Results and Discussion A total of 75 sessile species comprising scleractinian corals, algae, and sessile invertebrates were identified in die 36 transects sampled across the upper, middle and lower reef flat zones. 34 species of scleractinian corals from eight families were recorded (Table 1). The ma,onty of species were members of the family Merulinidae, represented by 18 species (or 53" o of all corals), followed by the Lobophyllidae and Poritidae, each represented > species (23% of all corals). All remaining 5 families were represented by 1 or 2 Fig. 1. Location of Nightcliff Reef off the north¬ eastern foreshore of Darwin Harbour. Coral communities in extreme environments Northern Territory Naturalist (2016) 27 87 species (24% of all corals). The more abundant corals, all with greater than 2% in mean percentage cover pooled for all 36 transects, were P/atygyra sinensis (Fig. 2), Pontes cf nigrescens, Astrea atria, Coelastrea aspera and Goniastrea retiformis. Secondary dominant corals (1-2%) were Po riles lit tea, Lobophyllia hemprichii, P/atygyra daedalea, Galaxea as treat a, Favites Fig. 2. Massive colony of P/alygyra sinensis at Nightcliff Reef, Darwin Harbour. This particular species and Coelastrea aspera have been sighted attaining colony sizes of up to 1.8 m in height. (Lawrance Ferns) 88 Northern Territory Naturalist (2016) 27 Ferns Table 1. Mean percentage cover of scleractinian corals, algae, other sessile invertebrates and physical substrates at Nightcliff Reef. Lifeform categories M = massive, S = submassivc, E = encrusting, D/C = digitate/coryombose, R = ramose, T - turf, BF = bladed foliosc, F = foliose, EC — erect coralline. Family Species Lifeform Upper Reef Flat (>1.4—1.8 m) Middle Reef Flat (>0.9-1.4 in) Lower Reef Flat (0.3-0.9 m) north central south north central south north central south CORALS Mcrulinidac Pbtygyra sinensis M 0.01 0.21 0.21 19.60 4.13 1.48 Ponndac Porites cf nierescens S/R 8.10 4.34 5.76 3.75 Mcrulinidac Astna curta M 0.06 2.85 0.15 1.66 3.93 3.25 2.35 Mcrulinidac Coelastrea aspera M 2.20 2.84 0.60 Mcrulinidac Goniastrra retiformis M 0.03 0.18 0.24 0.20 0.04 0.08 2.80 0.46 0.11 Poriddac Ponies In tea M 0.50 1.36 n/a Dead standing coral n/a 0.45 0.61 0.46 0.34 Lobophylliidae I j) bop fry Ilia hemprichu S 1.79 Mcrulinidac Platygpra daedalea M 0.23 1.19 0.26 0.11 Kuphylliidac Galaxea astreata E 0.13 1.24 0.25 Mcrulinidac l'antes abdita S 0.30 0.03 0.60 0.43 0.15 Mcrulinidac Drpastrea speciosa M 0.20 0.05 0.32 0.03 0.28 0.34 0.04 Mcrulinidac Dipastrea rotnmana M 0.23 0.48 0.14 0.09 n/a Jimmie corals n/a 0.11 0.08 0.09 0.06 0.32 0.16 0.09 Mcrulinidac Drpastrea matthaii M 0.15 0.33 0.08 0.04 Mcrulinidac Dipastrea amicomm M 0.03 0.13 0.09 0.29 0.04 Mcrulinidac Cyphdstrea serai/ia B 0.06 0.05 0.13 0.05 0.01 0.13 Mcrulinidac Dipastrea fonts M 0.12 0.04 0.17 0.03 0.05 Acroporidac Acropora millepora D/C 0.34 0.05 Mcrulinidac Dipastrea pallida M 0.01 0.10 0.16 0.04 Poriddac Pontes sp.1 M 0.24 0.03 Ponddac Goniopora cf. lobata S 0.04 0.14 0.09 Lobophylliidae Symphy/da recta M 0.01 0.23 Acroporidac Montipora enmtsdne E 0.13 0.06 Dcndrophyllidac Turbinaria mesentenna E 0.03 0.10 Dendrophyllidac Tnrbinana conspicua T 0.10 Mcrulinidac Merudna ampliata E 0.10 Sclcracrinia inccrtac scdis Ijeptastrea purpurea B 0.09 1 x>bophylUidae Ecbinophylba aspera E 0.09 Sclcracrinia inccrtac scdis Ijeptastrea transversa K 0.01 0.01 0.02 0.01 0.00 0.03 Mcrulinidac Cypbashrta rmcropbtbalma E 0.08 Mcrulinidac Goniastrea pectinate E 0.03 0.04 Continued on next page Coral communities in extreme environments Northern Territory Naturalist (2016) 27 89 Continued from previous page Family Species Lifcform Upper Reef Flat (>1.4—1.8 m) Middle Reef Flat (>0.9-1.4 m) Lower Reef Flat (0.3-0.9 ml north L-cntral south north cntral south north Merulinidac Cyphastrta chalcidicum E 0.05 Psammocoridac Psammocora continue! S 0.03 Merulinidac Goniastrea favulus M 0.03 I xjbophylliidae Aloseleya latistellata E 0.01 Subtotal: Corals 0.57 0.57 0.25 14.75 0.28 2.07 41.23 20.80 9.19 ALGAE Sargassaceae Sargassaceae 1 loldfasts BF 4.01 0.43 1.81 0.08 Dictyotaceae Padina australis F 12.55 20.39 6.84 0.03 0.63 0.15 0.03 Galaxauraceae Tric/eocarpa franilis T 0.16 0.61 0.69 0.23 0.01 Dictyotaceae Dictyopteris sp. BE 0.09 1.16 0.01 Corallinaceae Amphiroa fragilissima T 2.72 0.23 Sargassaceae Sargissopsis decurrens (Sighted only) BF Rhizophyllidaceac Portiena bornemannii T 1.21 0.24 0.05 0.30 0.20 Dictyotaceae Spatoglossum asperurn BE 0.32 1.24 Halimcdaccae 1 / a limeda cf tuna EC 0.01 0.01 0.12 Rhodomelaceae Pa/isada Perforata T 0.11 0.21 1.21 Rhodomelaceae II 3 11 T Cystocloniaceac Hypnea spinella T 0.28 0.08 Rhodomelaceae Acantbophora spici/era T 0.12 0.13 0.86 0.11 Gracilariaceae Gracilaria saiicornia T 0.28 0.33 0.61 Halimedaceae lialimeda opuntia EC 0.13 0.21 0.01 0.27 0.11 Anadyomenaceac Anadyomene plicata EC 0.22 0.21 0.07 0.26 0.33 Cystocloniaceac Hypnea talentiae T 0.24 0.10 0.45 Scytosiphonaceae Hydrodatbrus clatbratus T 0.08 0.51 0.13 Caulerpaceae Caulerpa lentillifera F 0.01 0.08 Rhodomelaceae Laurencia majuscula T 0.04 0.09 Rhodomelaceae Laurenda intricata T 0.12 0.05 0.17 Rhodomelaceae t'olypiodadia j£ T 0.01 0.01 0.22 0.09 Callithamniaceae Crouania attenuata T 0.01 0.03 0.29 Caulerpaceae Caulerpa racemosa E 0.03 Galaxauraceae Dicbotomaria ob/usata T Caulerpaceae Caulerpa serrulata E 0.04 Continued on next page 90 Northern Territory Naturalist (2016) 27 Ferns Continued from previous page Family Species Lifeform Upper Reef Flat (>1.4—1.8 m) Middle Reef Flat (>0.9-1.4 m) Lower Reef Flat (0.3-0.9 m) north central south north central south north central south Gclidicllaceae Gelidiella acerosa T 0.08 0.05 0.03 Rhodomclaceac Laurenda ohtusa T 0.05 0.04 Cystocloniaccac Hypnea pannosa T 0.02 0.03 0.02 Cystocloniaceac Hypnea cornufa T 0.02 Ulvaceac Viva intestinal'is F 0.03 0.02 Champiaccac Champia parvula T 0.04 Pithophoraccae Dictyosphaeria cavernosa F 0.04 Cystocloniaccac Hypnea if hamutosa T 0.03 Solieriaceae Sanonema filiforme T 0.01 0.01 Dichotomosiphonaccae Avrmnvilka erecta F 0.01 Dasycladaccac Neomeris annulata F 0.01 Dictyotaccac Dictyota bartayresiana (Sighted Only) BF Sub-total: Algae 20.05 26.06 11.82 0.05 6.87 1.12 0.03 0.05 SESSILE INVERTEBRATES Sponges 0.25 0.28 0.03 17.58 2.63 0.09 0.48 0.08 0.04 Other fauna (ascidians. soft corals) 0.03 0.04 0.01 BARK SUBSTRATES Coarse sand 2.84 11.79 57.40 5.26 3.46 4.06 Coral rubble 6.90 53 01 75 67 Muddv Rock 72.88 55.93 74.51 82.09 78.12 Muddy Sand 3.54 5.48 13.41 3.31 8.15 18 59 lotal f" o (.over) 100 100 100 100 100 100 100 100 100 abdita, Dipastrea .peciosa, and Dipastrea rotumana. Corals with ‘massive’ and ‘sub-massive’ lifeforms were the most common growth morphologies with 20 species (or 59% of all corals). 1 he other main lifeform was ‘encrusting’ comprising 12 species (35% total of a Cora ^‘ ^ <>nc S P CC * CS with digitate and tabulate lifeform categories were recorded (6% of all corals). Coral abundance and species nchness varied between vertical zones and horizontal ocalities across the reef. Physical factors such sediment type, micro-topography, standing ater, win the relative influence of each factor shifting between and within vertical zones and localities, is the likely cause of the observed variation (Table 1). Species cover a.u nchness increased markedly seaward from the upper reef flat zone (1.4-1.8 m) to the lower reef flat zone (0.3-0.9 m) which has the highest coral abundance and species nchness. I hisns consistent with similar studies that have found intertidal corals to be more successful below 1 m tidal height elevations and reaffirms the period of emersion me is an important factor determining the their upper vertical limit (Morrissey 1980; Coral communities in extreme environments Northern Territory Naturalist (2016) 27 91 Bull 1982). Coral abundance and species richness was significandy higher on the lower reef flat at the northern extent of the reef (41% mean cover, 26 species), and decreased towards the southern extent (9% mean cover, 13 species). This is attributed to a notable decrease in fine sediments at the northern extent, combined with a higher proportion of coral rock and rubble, and the presence of small tide pools with standing water (Table 1). High levels of fine sedimentation on substrates are not conducive to coral recruitment and growth (Rogers 1990; Rogers et al 1994). It is postulated that predominant north¬ westerly tidal movement drives local cross-shore transport of finer sediment from north to south. The coral species and lifeforms in these extreme conditions are characteristic of those that occupy sheltered reefs with high sedimentation and turbidity (Rosen 1971; Chappel 1980; Done 1982). Species with massive, submassive and encrusting life forms are more tolerant to thermal stress (Marshall & Baird 2000), while convex and steep sided growth morphologies facilitate sediment flow off surfaces (Lasker 1980; Rogers 1990). The most diverse species were from the Merulinidae, which are represented by species that are tolerant of thermal stress (Depczynski et al. 2013; Schoepf et al. 2015) and well defined association with turbid water and lengthy periods of emersion. Corals with large polyps, like most species from the Merulinidae are efficient at removing sediment (Marshall & Orr 1931), and species such as Dipastrea spp., Goniastrea spp, Lobop/g/lia hemphriebii, Aslrea curta, PJatygyra spp., and Sjmplryllia recta can easily manipulate silt and fine sands (Stafford- Smith & Ormond 1992). Species of the genera Coelastrea and Goniastrea, for example, are often found in conditions where corals may not be expected to survive (Vcron 1986) and are dominant in high latitude reefs in waters of high nutrients and turbidity (Fhomson & Frisch 2010). In the Arabian Gulf region, the Merulinidae has a greater representation of species compared to other coral families such as Acroporidae which has greater species diversity in the Indo-Pacific region (Coles 1993; Foster & Foster 2013). Species of the Acroporidae were scarce, with only two species represented at very low coral cover. The Acroporidae is reported to occur in other regions with high thermal stress (e.g. Craig et al 2001; Bauman et al 2013b; Dandan et al. 2015; Schocpf et al. 2015). At Nightcliff Reef it is probably further limited by the high turbidity' as it is a poor sediment rejector with small polyps less than 2 mm diameter (Stafford-Smith & Ormond, 1992). However, a small-polyped species, Pontes cf. nigrescens (Poritidae) was abundant on the middle and outer reef zones. This species has also been recorded as abundant amongst a coral community surveyed at Cobourg Peninsula, Northern Territory (Billyard 1995) and to the west on macro-tidal reef flats and lagoons off Sunday Island at the mouth of King Sound, Western Australia (Dobson 1999). In further regions, Parities nigrescens is reported as common on turbid fringing reefs to the south of Saudi Arabia (Jeddah to Jizan), but disappears to the north in clearer waters (Sheppard 1985). The species is regarded as a shallow water reef builder in environments with low light intensities and low wave energy (Cabioch et al. 1999). In northern Australia it appears to have adopted a similar niche to that occupied by Montipora digitata that is common on more sedimented 92 Northern Territory Naturalist (2016) 27 Ferns inshore reefs of eastern Australia, such as the reef flats of Magnetic Island, Queensland (Bull 1982; Mapstone et al. 1992). Sediments for the upper and middle reef flat zones across the majority of localities were predominantly muds and silts on rock with intervening patches of muddy sand. Benthic algae were both more abundant and more diversely represented in these zones and collectively contributed to 38 species (Table 1). The brown foliose species Paditta australis dominated the upper reef flat zone. Other conspicuous bladed foliose algae included Sargassacca holdfasts (probably Sargassopsis decurrens), Splatoglossum asperum and Dictyopteris sp. Considerable turf forming algae were also present on the upper reef flat zone, and dominated in the middle reef flat zone with Amphiroa jragilissitna , Portieria hontemanm, Trickocarpa fragilis. Acantbopbora specifera, Gracilatia salicoruia and Anadyomene plicata visually conspicuous and varying in relative abundance across localities. The lower reef flat zone, which was dominated by corals, exhibited minimal algal cover (0.03-0.05% mean cover, Table 1). There was no evidence of competition from algae with corals as described in eastern Australia (Morrisey 1980), the Arabian Gulf (Sheppard et al. 1992) and the Red Sea (Loya 1977b). It appears corals have a competitive advantage in this lower reef flat zone due to the greater water depth and ability to cope with the high turbidity which results in lower light availability to benthic algae. In the shallower upper tidal zones, longer and more frequent emerison times exclude most coral species, and this allows rapid colonisation of benthic algae which also gain improved light attenuation for photosynthesis. The algal species recorded at Nightcliff Reef have been widely reported across the Indo-Pacific, from Tanzania (Jassund, 1976) to eastern Australia (Morrissey 1980; Ngan & Price 1980). The abundance of benthic algal species at Nightcliff Reef is likely to be seasonal with species exhibiting variable growth and dominance between wet and dry seasons (e.g. Benayahu & I -ova 1977b; Lawson & John 1987; Vuki & Price 1994). The dominance of fleshy brown algae in the upper reef flat zone, with the co-occurrence of turfing algae is similar to the reef flat zonation described in the Gulf of Eilat, Red Sea (Benayahu & Loya 1977a, 1977b). A notable observation was the limited representation of erect coralline algae from the genus Halimeda which is an important contributor to calcium carbonate accretion and grows abundantly on reef flats in eastern Australia (Morrissey 1980), Guam (Merten 1971) and the Gulf of Mannar, India (Rao 1972). However, in similar extreme environmental conditions such as the .Arabian Gulf region, Halimeda is not recognised as a major component of inshore reefs (Sheppard 1985; Sheppard et al 1992). The coral species of Darwin Harbour offer valuable insights to physiological and evolutionary adaptations to persist in the most extreme environmental conditions. It is evident from this study and comparative investigations elsewhere that species with large polyps, thick skeletal tissue and massive or submassive lifeform strategics are amongst the most successful to surviving high sedimentation, nutrients and water temperatures. Coral communities in extreme environments Northern Territory Naturalist (2016) 27 93 The close proximity of Nightcliff Reef to coastal development, land-based sources of pollution and marine development combined with the changing climate pose future conservation challenges for this reef which is currently deemed at high risk to integrated local threatening processes (Burke et al 2011). Acknowledgments I am grateful to the Parks and Wildlife Commission of the Northern Territory for funding, and support from Michael Guinea, Ron Billyard, lan Pulton, Belinda Oliver, Ian Price and John Collins. This research was conducted towards partial fulfillment of a Bachelor of Honours degree at the Charles Darwin University. References Australian Bureau of Meteorology (2013) Record sea surface temperatures, (accessed 15 November 2013). Bauman A.G., Baird A.H. and Cavalcante G.H. (2011) Coral reproduction in the world’s warmest reefs: southern Persian Gulf (Dubai, United Arab Emirates). Coral Reefs 30, 405-413. Bauman A.G., Pratchett M.S., Baird A.H., Riegl B., Heron S.K and Feary D.A. (2013a) Variation in the size structure of corals is related to environmental extremes in the Persian Gulf. Marine Environmental Research 84, 43—50. Bauman A.G., Feary D.A., Heron S.F., Pratchett M.S. and Burt J.A. (2013b) Multiple environmental factors influence the spatial distribution and structure of reef communities in the northeastern Arabian Peninsula. Marine Pollution bulletin 72, 302—312. Benayahu Y. and Loya Y. (1977a) Space partitioning by stony corals, soft corals and benthic algae on the coral reefs of the northern Gulf of Eilat (Red Sea). I ielgolander Wissenschlftliche Meeresuntersuchungen 30, 362—382. Benayahu Y. and Loya Y. (1977b). Seasonal occurrence of benthic-algae communities and grazing regulation by sea urchins at the coral reefs of Eilat, Red Sea. In: Proceedings of Third International Cora! Reef Symposium Volume 1: biology (ed. Taylor D.J..), pp. 383-390. Rosenstiel School of Marine and Atmospheric Science, Miami, Florida. Billyard R.W. (ed) (1995) The Cultural and Ecological Significance of Popham Creek, Cobourg Peninsula. Parks and Wildlife Commission of the Northern Territory, Darwin, Australia. Budd A.F., I'ukami 11., Smith N.D. and Knowlton N. (2012) Taxonomic classification of the reef coral family Mussidae (Cnidaria: Anthozoa: Scleracdnia). Zoologicaljournal of the Unman Society 166, 465-529. Burke L. M., Rcytar K., Spalding M. and Perry, A. (2011) Reefs at Risk Revisited. World Resources Institute, Washington, DC. Cabioch G., Camoin G. and Montaggioni L. (1999) Postglacial growth history of a French Polynesian barrier reef tract, Tahiti, central Pacific. Sedimentology 46, 985-1000. Chappell J. (1980) Coral morphology, diversity and reef growth. Nature 286, 249-252. Chiappone M. and Sullivan K.M. (1991) A comparison of line transect versus linear percentage sampling for evaluating stony coral (Scleractinia and Milleporina) community similarity and area coverage on reefs of the central Bahamas. Coral Reefs 10, 139-154. Coles S. L. (1997) Reef corals occurring in a highly fluemating temperature environment at Fahal Island, Gulf of Oman (Indian Ocean). Coral Reefs 16, 269-272. Coles S. L. (2003) Coral species diversity and environmental factors in the Arabian Gulf and the Gulf of Oman: a comparison to the Indo-Pacific region. Atoll Research bulletin 507, 1-19. 94 Northern Territory Naturalist (2016) 27 Ferns Cribb A.B. (1983) Marine Algae of the Southern Great Barrier Reef. Part 1. RJmdophyta. Australian Coral Reef Society Handbook No. 2, Brisbane. Cribb A.B. and Cribb J. W. (1985) Plant Life of the Great Barrier Reef and Adjacent Shores. University of Queensland Press, St. Lucia, Australia. Craig P., Birkeland C. and Belliveau S. (2001) High temperatures tolerated by a diverse assemblage of shallow-water corals in American Samoa. Coral Reefs 20,185-189. Dandan, S.S., Falter, JX., Lowe, R.J. and McCulloch, M.T. (2015) Resilience of coral calcification to extreme temperature variations in die Kimberley region, northwest Australia. Cora! Reefs 34, 1151-1163. Depczynski M., Gilmour J. P., Ridgway T., Barnes H. et at. (2013) Bleaching, coral mortality and subsequent survivorship on a West Australian fringing reef Cora! Reefs 32, 233-238. Dobson G. (1999) Assessing and limiting predation and other sources of mortality in reseeded Trochus niloricus. Unpublished MSc Thesis. Northern Territory University, Darwin. Done TJ. (1982) Patterns in the distribution of coral communities across the central Great Barrier Reef. Coral Reefs 1, 95—107. English S., Wilkinson C. and Baker V. (eds.) (1994) Survey Manual for Tropical Marine Resources. Australian Institute of Marine Science, Townsville. Foster K.A. and Foster G. (2013) Demography and population dynamics of massive coral communities in adjacent high latitude regions (United Arab Emirates). PLoS ONE 8, e71049. Gilmour J.P., Cooper T.F., Fabricius K.E. and Smith L.D. (2006) Early Warning Indicators of Change in the Condition of Corals and Coral Communities in Response to Key Anthropogenic Stressors in the Pilbara, Western Australia. Australian Institute of Marine Science, Western Australia. Glynn P.W. (1993) Coral reef bleaching: ecological perspectives. Coral Reefs 12, 1-17. Guiry M.D. and Guiry G.M. (2013) AlgaeBase. World-wide electronic publication. National University of Ireland, Galway, (accessed 12 December 2015). Hooper j.N.A. (1987). Structural features of the bendiic community of East Point Fish Reserve. A comparative study between oceanic, nearshore and inshore reefs of Northwest Australia. In: Proceedings of a workshop on research and management held in Darwin, 2-3 September, !9H7 (eds Larson H.K., Michie M.G. and Hanley J.R.), pp. 215-225. Australian National University, North Australia Research Unit, Darwin. Huang D., Benzoni F., Fukami H., Knowlton N., Smith N. D. and Budd A. F. (2014) Taxonomic classification of the reef coral families Merulinidae, Montastraeidac, and Diploastracidac (Cnidaria: Anthozoa: Scleractinia). Zoological Journal of the Linnean Society 171, 277-355. Hughes T. R, Baird A. H., Bellwood D. R., Card M. et at. (2003). Climate change, human impacts, and the resilience of coral reefs. Science 301, 929—933. Jaasund E. (1976) intertidal Seaweeds in Tanzania: A lie/d Guide. University of Tromse Press, Norway. Lasker H.R. (1980). Sediment rejection by reef corals: the role of behaviour and morphology in Montastrea cavernosa (Linnaeus). Journal of Experimental Marine Biology and Ecology 47, 77—87. Lawson G.W and |ohn D.M. (1987) 7 he Marine Algae and Coastal Environment of Tropica / West Africa. J. Cramer Publisher, Berlin. Mapstone B.D., Choat |.l 1., Cumming R.L. and (ixlcy W.G. (1992) The I ringing Reefs of Magnetic Island: Benthic Biota and Sedimentation - a Baseline Study. Great Barrier Reef Marine Park Authority Research Publication No. 13, Townsville. Marshall P.A. and Baird A.I1. (2000) Bleaching of corals on the Great Barrier Reef: differential susceptibilities among taxa. Cora! Reefs 19, 155—163 Marshall S.M. and Orr A.P. (1931) Sedimentation on Low Isles Reefs and its relation to coral growth. Scientific Report of the Great Barrier Reef Expedition 3, 273—312. Coral communities in extreme environments Northern Territoty Naturalist (2016) 27 95 Merten M.J. (1971) Ecological observations on Halimeda macrotoba Decaisne (Chlorophyta) on Guam. Micronesia 7,27-44. Michie M.G. (1987) Distribution of Foraminifera in a macrotidal tropical estuary. Port Darwin, Northern Territory of Australia. Australian Journal of Marine and Freshwater Research 38 249-259. Morrissey J. (1980) Community structure and zonation of macroalgae and hermatypic corals on a fringing reef flat of Magnetic Island Queensland, Australia. Aquatic liolany 8, 91-139. Ngan Y. and Price L.R. (1980) Distribution of intertidal benthic algae in the vicinity of Townsville, Tropical Australia. Australian Journal of Marine and Freshwater Research 31, 175—191. Price I.R. and Scott FJ. (1992) The Turf Algal Flora of the Great Barrier Reef. Part 1 Rhodoplyla. James Cook University of North Queensland Press, Townsville. Rao M. (1972) Coral reef flora of the Gulf of Mannar and Palk Bay. In Proceedings of the 1st International Symposium on Corals and Cora! Reefs (eds Mukundan M. and Gopinadha Pillai C.S.), pp. 217—230. Marine Biological Association of India, India. Riegl B. (1999) Corals in a non-reef setting in the southern Arabian Gulf (Dubai, UAE): fauna and community structure in response to recurring mass mortality. Cora! Reefs 18, 63—73. Riegl B. and Purkis S. (eds) (2012) Cora! Reefs of the Gulf Adaptation to Climatic Extremes. Coral Reefs of the World Volume 3, Springer, Netherlands. Rogers C.S. (1990) Responses of coral reefs and reef organisms to sedimentation. Marine Ecology Progress Series 62, 185-202. Rogers C.S., Fitz H.C., Gilnaek M., Beets J. and Hardin J. (1984) Scleractinian coral recruitment patterns at Salt River Submarine Canyon. St. Croix, U.S.V.I. Coral Reefs 3, 69-76. Rosen B.R. (1971) Principal features of reef coral ecology in shallow water environments of Mahe, Seychelles. Symposium oj the Zoological Society of London 28, 163—183. Schoepf V., Stat M., Falter J.L. and McCulloch M. (2015) Limits to thermal tolerance of corals adapted to a highly fluctuating, naturally extreme environment. Science Report 5, Article 17639. Semeniuk V. (1985) Mangrove environments of Port Darwin, Northern Territory: the physical framework and habitats. Journal of the Royal Society of Western Australia 67, 81-97. Sheppard C. and Sheppard A. (1991) Corals and coral communities of Arabia. Fauna of Arabia 12, 3-17. Sheppard C.R.C., Price A. and Roberts C. (1992) Marine Ecology of the Arabian Region. Patterns and Processes in Extreme Tropical Environments. Academic Press, London. Stafford-Smith M.G. and Ormond R.F.G. (1992) Sediment-rejection mechanisms of 42 species of Australian scleractinian corals. Australian Journal of Marine and Freshwater Research 43, 683-705. Sullivan K.M. and Chiappone, M. (1992) A comparison of belt quadrat and species presence/ absence sampling of stony coral (Sclcractinia and Millcporina) and sponges for evaluating species patterning on patch reefs of the central Bahamas. Bulletin of Marine Science 50,464—488. VeronJ.E.N. (1986) Corals of Australia and the halo-Pacific. Angus and Robertson, Australia. VeronJ.E.N. and Pichon M. (1976) Sckractinia of Eastern Australia. Part I. Families Thamnasteriidae, Astrocoeniidae, Poalloporidae. AIMS Monograph Series 1, Australian Institute of Marine Science, Townsville. VeronJ.E.N. and Pichon M. (1980) Sckractinia of Eastern Australia. Part III. Families Agaraciidae, Siderastreidae , I'ungiidae, Ocutinidae, Merulinidae, Mussidae, Pectiniidae, Caryoplrylliidae, Dendrophylliidae. AIMS Monograph Series 4, Australian Institute of Marine Science, Townsville. VeronJ.E.N. and Pichon M. (1982) Sckractinia of Eastern Australia. Part 1 1 I amities Poritidae. AIMS Monograph Series 5, Australian Institute of Marine Science, Townsville. 96 Northern Territory Naturalist (2016) 27 Fcrn s Veron J.E.N., Pichon M. and Wijsman-Best M. (1977) Scleractinia of Eastern Australia. Part If Families Faviidae , TradryphyUidae. AIMS Monograph Series 3, Australian Institute of Marin e Science, Townsville. Veron J.E.N. and Wallace C.C. (1984) Scleractinia of Eastern Australia. Part I i Family Acroporidae. AIMS Monograph Series 6, Australian Institute of Marine Science, Townsville. Yuki V.C. and Price I.R. (1994) Seasonal changes in the Sargassum populations on a fringing reef Magnetic Island, Great Barrier Reef Region, Australia. Aquatic botany 48, 153—166. Wolstenholme J., Dincsen Z.D. and Aldcrslade P. (1997) Hard corals of the Darwin region. Northern Territory, Australia. In: Proceedings of the Sixth International Marine biological Workshop; The Marine P/ora and Fauna of Darwin Harbour, Northern Territory, Australia (ed. Hanley R) pp. 381—398. Museum and Art Gallery of the Northern Territory and the Australian Marine Sciences Association, Darwin. WoRMS Editorial Board (2015) World Register of Marine Species, (accessed 24 November 2015). Wynne M.J. (2011) The benthic marine algae of the tropical and subtropical Western Adantic: changes in our understanding in the last half century. Algae 26, 109-140. Northern Territory Naturalist (2016) 27: 97—101 Short Note First record of two mangrove leaf slugs, Elysia leucolegnote and E. bangtawaensis (Sacoglossa: Plakobranchidae), in mangrove forests in the Northern Territory Adam J. Bourke 1 , Carmen Walker 1 and Richard C. Willan 2 1 EcoScience NT, 29 Ostermann St, Coconut Grove, Darwin, NT 0810, Australia Email: ccoscicncc2@bigpond.com 2 Museum and Art Gallery of the Northern Territory, GPO Box 4646, Darwin, NT 0810, Australia Abstract Here we report for the first time on the occurrence of the distinctive and highly ephemeral sap-sucking sea slugs Elysia leucolegnote and E. bangtawaensis from mangrove forests from Darwin Harbour, Northern Territory, Australia. Individuals of both species apparently attain smaller body size than their counterparts elsewhere in Australia and the Indo-Pacific region, with maximum extended crawling lengths recorded between 17-22 mm. It appears the northern Australian (i.e. Northern Territory and northern Queensland) populations of E. bangtawaensis differ consistently from their counterparts elsewhere in the world in aspects of (parapodial and rhinophoral) colouration. The ability to retain functioning chloroplasts sequestered from algal host(s) is widespread in the group of sea slugs known as sap-sucking slugs (order Sacoglossa). Numerous species in the genus Elysia (family Plakobranchidae, the largest family numerically in the Sacoglossa) are known for their ability to sequester live chloroplasts within their extensively branched digestive diverticula, imparting a bright green colour (c.g. Trench et at. 1973; llumpho et al 2008; Jesus et al 2010). Some of them match their host food precisely (Burn 1998). The genus Elysia is species-rich, with some 95 named species (Bouchct & Gofas 2015) and at least that number again undescribed (RCW pers. obs). Its members are mostly small in adult size (< 20 mm) and live on algae in tidal pools, seagrass meadows or in sublittoral algal beds (Swennen 1997). However, four comparatively large (25-50 mm) species are specific to mangrove habitats, aggregating on mud in shaded pools without any apparent algae in the immediate vicinity (Swennen 2011). These mangrove-dwelling species, which have all been described in the last 25 years, are Elysia leucolegnote , E. bangtawaensis , E. singaporensis and E. benga/ensis (Swennen 2011). Collectively, these four species arc commonly called mangrove leaf slugs, as animals resemble a fallen mangrove leaf (in both shape and coloration) when they have their wing-like parapodia relaxed and fully extended (by contrast, all other species of Elysia have the parapodia folded up on the dorsal side, with the margins meeting more or less in a wavy line mid-dorsally). All these mangrove leaf slugs are restricted to the 98 Northern Territory Naturalist (2016) 27 Bourke et al tropical Indo-west Pacific region, but only E. leucolegnote and E. bangtawaensis occur in both hemispheres, ranging from the western coast of India to the north-eastern coast of Australia (Swennen 1997; Rudman 2007; Rudman 2009; Swennen 2011). Despite the wide distribution and presence of both species in eastern Australia, neither has been previously recorded from anywhere in the Northern Territory. Here we document the first occurrence of E. leucolegnote and E. bangtawaensis in Northern Territory mangroves, specifically Darwin Harbour. Like most others throughout the world, the discoveries were of hundreds of individuals, not just one or two. During the early dry season (April) of 2014 approx. 115 adults and juveniles of E. leucolegnote were encountered, and during the dry season of the following year (April-May 2015) around 90 adults of E. bangtawaensis were encountered. Elysia leucolegnote (Figs 1—3) On 14 April 2014 approx. 75 adults and juvenile E. leucolegnote were found in a puddle in the landward mangrove fringe at Bayview, Sadgroves Creek (12.4419°S, 130.8611°E). These individuals were identified by AJB and confirmed by RCW. Elysia leucolegnote is characterised by having a white or yellowish border to the parapodia, a distinctive white triangular mark on the head, and a white line over the dorsal side of each rhinophore that connects with the one from the other rhinophore on top of the head (Swennen 2011) (Fig. 1). All Darwin animals displayed these characteristic features. Living individuals ranged in size from less than 4 mm up to 22 mm in extended crawling length (ECL) and each had a yellowish-green coloured digestive gland (Fig. 2), indicating they had not fed for some months (Swennen 2011). Further specimens were found on 20 April when approx. 40 small (10—15 mm) dark green coloured slugs were recorded in shallow puddles in a low, closed Stilt-root Mangrove (Rbiqopbora sty/osa) forest at Virginia, Elizabeth River (12.5690°S, 131.0156°E) (Fig. 3). All the E. leucolegnote observed in Darwin mangroves were notably smaller than the maximum size of 41 mm known for this species elsewhere (Swennen 2011). These records of Elysia leucolegnote from Darwin Harbour can be added to those already known for this species from Australia - from northern New South Wales and southern Queensland (Allan 1950; Thompson 1973; Burn 1998). However, it is possible some (if not all) of these records might relate to E. bangtawaensis as there are no recent records of E. leucolegnote from this region (RCW pers. obs.). Its occurrence in the Kimberley region is to be expected. Elysia bangtawaensis (Figs 4—7) Elysia bangtawaensis is characterised by the prominent reddish or orange glandular spots along its parapodial margin (Swennen 2011). 1 he description of its colour in life stated “no epidermal pigmentation other than the red and white glands” (Swennen 1998) and noted that the tips of rhinophores were pale-coloured or orange (Swennen 2011). This species was first recorded in the Northern Territory on 21 April 2015, when approx. 20 small (13 mm) individuals were found in shallow, party shaded puddles amongst Mangrove leaf slugs in Darwin Harbour Northern Territory Naturalist (2016) 27 99 1 Figs 1-3. Iilysiu kucolegnote. Specimens photographed in the laboratory and in situ. 1. An individual with partially relaxed parapodia (scale bar = 5 mm); 2. Yellowish-coloured, starved individuals with fully relaxed parapodia congregated in a shallow puddle near the landward mangrove edge; 3. Dark-green individuals congregated in a shallow puddle (4 mm depth) in a Rhi^opbora sty/osa forest. (Adam Bourke) Pornupan Mangrove ( Sonne ratio alba ) pncumatophores in the seaward mangrove community (Fig. 4) about 120 m east of the East Arm Boat Ramp carpark (12.4842°S, 130.9132°E). These individuals were identified by RCW. One month later (i.e. on 20 May) approx. 70 small slugs (9—17 mm) were observed in a similar habitat (Fig. 5) about 240 m west of the East Arm Port Precinct (12.4831°S, 130.9239°E). These records of Efysia bangtawaensis from Darwin Harbour can be added to those already known for this species from Australia — from northern New South Wales (Cobb 2007, which were also first identified by RCW; Riek 2015), and northern Queensland (Mitchell 2009). As with H/ysia kucolegnote , its occurrence in the Kimberley region is to be expected. All the / '.tysia bangtawaensis individuals recorded from Darwin I larbour differed from previously described specimens in having black tips to their rhinophores (big. 6) and microscopic, iridescent blue spots covering much of the dorsal and ventral margin of the parapodia (Fig. 7). Interestingly, E. bangtawaensis with tiny, metallic blue spots have also been recorded from mangroves in Cairns, northern Queensland (Mitchell 2009). 100 Northern Terriloiy Naturalist (2016) 27 Bourke et al. Fig. 4. Efysia bangtawciensis. Individuals with relaxed parapodia displaying characteristic leaf¬ shaped bodies (Adam Bourke). Fig. 5. /:. bangUiwaemis aggregated in a shallow puddle in the seaward mangrove community. (Adam Bourke) However, to our knowledge animals with black-tipped rhinophores have not been reported previously. As is the case with E. leucolegnote , all individuals of E. bangtawaensis from Darwin Harbour were notably smaller than the maximum live length of 52 mm known for this species (Swennen 2011). That both of these large species of Efysia are highly ephemeral in time and space is shown by the fact that we systematically sampled in mangrove forests in Darwin Fig. 6. A Darwin E. bangtanwnsis specimen displaying black-tipped rhinophores and characteristic orange glandular dots along the parapodia! border (scale bar — 5 mm) (Adam Bourke). Fig. 7. Magnified view of E. bangtawaensis specimen from Darwin displaying tiny, iridescent blue speckles scattered along the parapodial border. (Adam Bourke) Mangrove leaf slugs in Darwin Harbour Northern Territory Naturalist (2016) 27 101 Harbour for 25 years and never encountered them previously. Therefore, their detection is unlike that of the caddis slug Aiteng sp., which remained undetected in the harbour because it is microscopic (Neusser et al 2015) . Acknowledgements Sincere thanks to Kces (Cornelis) Swcnnen for responding to numerous enquiries regarding the distribution and external features of both species. Kristin Metcalfe kindly offered comments on an early draft of this paper References Allan J.K. (1950) Australian Shells: with related animals living in the sea, in freshwater and on the land. Georgian House, Melbourne. Bouchet P. and Gofas, S. (taxonomic editors) (2015). E fysia Risso, 1818. In: MolluscaBase (2015). Accessed through: Wodd Register of Marine Species at (accessed 29 October 2015) Burn R. (1998) Order Sacoglossa. In: Molhtsca: The Southern Synthesis: Fauna of Australia Volume 5 (eds Becsley, PI.., Ross, G.J.B. and Wells, A.E.), pp. 961-974. CSIRO Publishing, Canberra. Cobb G. (2007) Efysia bangtawaensis in nthn New South Wales. [Message in] Sea Slug Forum. Australian Museum, Sydney, (accessed 29 October 2015) Jesus B., Ventura P. and Calado G. (2010) Behaviour and a functional xanthophyll cycle enhance photo regulation mechanisms in the solar-powered sea slug Efysia timida (Risso, 1818). Journal of Experimental Marine Biology and Ecology 395(1), 98-105. Mitchell A.C. (2009) Efysia bangtawaensis from Cairns, nthn Queensland. (Message in] Sea Slug Forum. Australian Museum, Sydney, (accessed 20 July 2015) Neusser T.P., Bourkc A., Metcalfe K. and Willan R.C. (2015) First record of Aitengidac (Mollusca: Panpulmonata: Acochlidia) for Australia. Northern Territory Naturalist 26, 27-31. Rick D.W. (2015) Efysia bangtawaensis. Sea slugs and other marine invertebrates of the Tweed-Byron coast, Australia, (accessed 29 October 2015) Rudman W.B. (2007) Comment on Efysia bangtawaensis in nthn New South Wales by Gary Cobb. |Messagc in] Sea Slug Forum. Australian Museum, Sydney, (accessed 20 July 2015) Rudman W.B. (2009) Comment on Efysia bangtawaensis from Cairns, nthn Queensland by Andrew Mitchell. (Message in] Sea Slug Forum. Australian Museum, Sydney. (accessed 20 July 2015) Rumpho ALE., Worful J.M., Lee J., Kannan K. et at. (2008) Horizontal gene transfer of the algal nuclear gene psbO to the photosynthetic sea slug Efysia ch/orotica. Proceedings of the National Academy of Sciences 105 (46), 17867-17871. Swenncn C. (1998) Two new gastropods, Efysia bangtawaensis and E. siamensis from southern Thailand (Opisthobranchia, Sacoglossa, F.lysiidae). Bulletin Zoo/ogisch Museum 16(6), 33—39. Swcnnen C. (2011) Large mangrove-dwelling Efysia species in Asia, with descriptions of two new species (Gastropoda: Opisthobranchia: Sacoglossa). Raffles Bulletin of Zoology 59, 29-37. Thompson T.E. (1973) Sacoglossan gastropod molluscs from eastern Australia. Proceedings of the Ma/aco/ogicaf Society of Australia 40, 239-251. Trench R.K., Boyle |.E. and Smith D.C. (1973) The association between chloroplasts of Codium fragile and the mollusc Efysia tnridis. 11. Chloroplast ultrastructurc and photosynthetic carbon fixation in E. viridis. Proceedings of the Royal Society of latndon B: Biological Sciences 184 (number 1074), 63-81. Northern Territory Naturalist (2016) 27: 102-105 Short Note Field identification of the Platevindex mangrove slugs (Mollusca: Gastropoda: Onchidiidae) of Darwin Harbour Adam J. Bourke Ecoscience NT, 29 Ostermann Street, Coconut Grove, NT 0810, Australia Email: ecoscience2@.bigpond.com Abstract Darwin Harbour supports nine species of mangrove slugs (family Onchidiidae) and currendy the names of all them are unknown. It appears that the characters distinguishing the two Platevindex species allow animals to be accurately identified in the field on the basis of external characters and on differences in habitat. This note provides descriptions and information on the external characteristics of the two Platevindex species. Mangrove slugs as they are commonly known, comprise a family of shell-less, pulmonate (air-breathing) gastropod molluscs. Members of the Onchidiidae arc particularly special ecologically as they constitute one of only four families of gastropods (out of more than 400 families) with species living in marine, brackish, freshwater and terrestrial habitats (Dayrat 2009a). Onchidiid slugs have a wide geographic distribution, but most genera arc exclusively found within the tropical and subtropical Indo-West Pacific region (Dayrat 2009b). The Onchidiidae is a poorly-known family with its systcmatics in a state of confusion (Dayrat 2009b, Dayrat, Zimmermann & Raposa 2011). Animals in this family have been understudied since the last experts specialising in the taxonomy of the family were active more than 70 years ago and there is currendy no expert able to reliably identify members of the Onchidiidae (Dayrat 2009a; Benoit Dayrat pers. comm. 2013). At present, five genera of onchidiids are thought to inhabit the mangroves of Darwin Harbour, only two of which have been assigned names — Platevindex and Peronia (Benoit Dayrat & Irish Goulding, pers. comm. 2016). Among these genera, nine species have been identified, however none has been formally documented and the nomcnclatural status of all of them remains unknown. Hence, scientists dealing with them are required to allocate operational taxonomic units (OTUs, or ‘working names’) to these unnamed species until they arc documented by a taxonomic specialist. This note provides descriptions and information on the external morphological characteristics (i.e. shape, structure, colour and pattern) of the two unnamed species of / latevindex — herein referred to as Platevindex sp. 1 (blue) and Platevindex sp. 2 (orange). / lata index is a common tree-climbing genus of onchidiid inhabiting mangrove forests tie Darwin region of the Northern Territory and the two species arc frequently c >untered on the trunks and branches of mangrove trees during low tides. The genus Field identification of Platevindex slugs Northern Territory Naturalist (2016) 27 103 is characterised externally by having a noticeably narrower foot than underside of dorsal surface (= hyponotum) (i.e., a ratio of 0.25-0.3 foot width to hyponotal width depending on the degree of body contraction) (see Figs 1, 3), in contrast to other onchidiids. Field identification of Darwin’s Platevitidex slugs Most onchidiids cannot be easily identified by non-specialists as few species display distinctive external traits (Dayrat 2010). It is the author’s opinion, however, that the unique external features and colouration of Darwin Harbour’s two Platevindex slugs do allow them to be distinguished and identified accurately to species in the field. The following descriptions of the external characteristics plus photographs of living animals of Platevindex sp. 1 (blue) and Platevindex sp. 2 (orange) are presented as an aid to distinguishing between them in the field. The characteristics provided here are based on living adult individuals. Caution must be exercised when identifying juveniles though, as animals may differ in shape, colour and pattern as they mature. Where notable differences in external morphology or colour are evident between adults and juveniles, descriptions and photographs are provided. Platevindex sp. 1 (blue) (Figs 1, 2) Description: Size to 43 mm in extended body length (pers. obs.). Body ovate-elongate in shape, juveniles and subadults more circular; distinctively flattened. Notum (= dorsal surface) leathery, either smooth or warty in appearance, always moist to the touch. Photoreceptors (dorsal eyes) single, present almost to edge of notum. Foot sole distinctively narrow. Colour of notum variable, but usually dark grey with lighter grey- brown mottled patches or broken bands, commonly with dark black-grey reticulated motding around edge; mottling weaker in adults, but conspicious in juveniles and subadults (Fig. 2). Head, oral lobes and tentacles usually darker than underside of body. Foot sole commonly yellowish to light or dark cream in colour. Hyponotum (= underside Fig. 1. Dorsal and ventral views of living Fig. 2. Subadult Platevindex sp. 1 (blue) displaying adult Platevindex sp. 1 (blue). The specimen the distinctive reticulated dark mottling around is 40 mm in extended crawling length. the edge of the notum. The specimen is 24 mm in (Adam Bourke) extended crawling length. (Adam Bourke) 104 Northern Territory Naturalist (2016) 27 Bourke of dorsal surface) always pale bluish grey colour in adults, never with a motded dark colour radiating outwards from foot sole; juveniles commonly lighter cream in colour. Remarks on preserved specimens: Disregarding contraction of the body and some loss of colouration, most of die diagnostic features described above are clearly distinguishable for this species in preserved specimens (i.e. those preserved in 70% ethanol). In particular, the lack of dark mottling surrounding the foot sole remains distinct in preserved specimens. Ecology: This species may occur throughout the entire mangrove forest (pers. obs.), but adults are most commonly encountered on the trunks and branches of trees within forests dominated by Stilt-root Mangrove {Rhispphora stylosa) and Pornupan Mangrove {Sonneratia alba). Individuals arc regularly found feeding on large woody debris in the more seaward mangrove forest zones, but rarely observed on the forest floor (pers. obs.). Global distribution: The true extent of this species is uncertain. Currently it is only known definitely from Halmahera, the Maluku Islands, Queensland and the Northern Territory (Benoit Dayrat pers. comm. 2016). Platevindex sp. 2 (orange) (Fig. 3) Description: Size to 55 mm in extended body length (pers. obs.). Body ovate- elongate in shape, juveniles and subadults more circular; distinctively flattened. Notum (= dorsal surface) leathery, commonly having a very warty' appcrance resulting from laterally arranged raised bumpy ridges; usually dry to the touch. Foot sole distinctively narrow. Photoreceptors (dorsal eyes) single, never present at edge of notum, Colour of notum dark brown to brown. Head, oral lobes and tentacles rarely darker than underside of body. Foot sole always distinctively orange colour in adults; juveniles commonly pale to dark brown. Hyponotum (= underside of dorsal surface) yellowish-orange in colour, always with a mottled dark colour radiating outwards from foot sole. Remarks on preserved specimens: Disregarding contraction of the body and some loss of colouration, most of the diagnostic features described above are clearly distinguishable for this species in preserved specimens (i.e. those preserved in 70% ethanol). In particular, the presence of dark mottling radiating outwards from the foot sole is distinctive and remains distinct in preserved specimens. K - *■ • ' K • f £ , \v t \\ !te ■ . •: vtv - i: ' ' f \y."\ . i ]: , V'? l \ £$l Ip - its {' f - . Fig. 3. Dorsal and ventral views of living adult Platevindex sp. 2 (orange). The specimen is 47 mm in extended crawling length. (Adam Bourke) Field identification of Platevindex slugs Northern Territory Naturalist (2016) 27 105 Ecology: Platevindex sp. 2 is restricted to landward mangrove communities dominated by Smooth-fruited Spur Manrgove (Ceriops australis) in the upper intertidal (pers. obs.). Indiviuals are commonly observed feeding on the trunks and buttress roots of Ceriops australis trees and after spring high tides and rain, individuals are active on the forest floor (pers. obs.). During the dry season, slugs aestivate in the sediment, taking refuge inside crab burrows and crevices at the bases of trees (pers. obs.). Global distribution: Philippines, throughout the whole of Malaysia, Northern Territory (Benoit Dayrat pers. comm. 2016). Acknowledgements The author thanks Kristin Metcalfe, Richard Willan and Benoit Dayrat for reviewing earlier versions of this article. References Dayrat B. (2009a) Onchidiidae Diversity & Evolution. Benoit Dayrat Laboratory, School of Nauiral Sciences & Sierra Nevada Research Institute. University of California at Merced. (accessed 20 )uly 2014). Dayrat B. (2009b) Review of the current knowledge of the systematics of Onchidiidae (Mollusca: Gastropoda: Pulmonata) with a checklist of nominal species. Zootaxa 2068, 1-26. Dayrat B. (2010) Comparative anatomy and taxonomy of Onchidium vaigense (Gastropoda: Pulmonata: Onchidiidae). Mo/Juscan Research 30(2), 87—101. Dayrat B., Zimmermann S. and Raposa M. (2011) Taxonomic revision of the Onchidiidae (Mollusca: Gastropoda: Pulmonata) from the tropical Eastern Pacific .Journal of Natural History 45, 939-1003. Northern Territory Naturalist (2016) 27: 106—120 Research Article Captain King’s lost weevil - alive and well in the Northern Territory? Stefanie K. Oberprieler 123 , Debbie Jennings 4 and Rolf G. Oberprieler 4 1 CSIRO Tropical Ecosystem Research Centre, PMB 44 Winnellie, Darwin, NT 0822, Australia Email: sicf.oberpridcrthksiro.au 2 Research Institute for the Environment and Livelihoods, Charles Darwin University, Darwin, Darwin, NT 0909, Australia 3 Research School of Biology, Australian National University, Acton ACT 0200, Canberra, ACT 2601, Australia 4 CSIRO, Australian National Insect Collections, GPO Box 1700, Canberra, ACT 2601, Australia Abstract The discovery of a ‘hairy’ yellow weevil in Kakadu National Park in 1995, akin to a widely distributed pest species of agricultural crops in South-East Asia (but not Australia), the so-called ‘Gold-dust Weevil’ (Ilypomeces ‘squamosa?), prompted us to investigate the taxonomy and distribution of this weevil in order to determine the identity and origin of the Kakadu specimen. The ‘Gold-dust Weevil’, whose correct scientific name is H. pttlinger (Herbst, 1795), is a sexually dimorphic and variable species and has been described under various names in the literature, but its taxonomy and nomenclature have never been investigated. The results of our research to date indicate that it comprises a complex of closely similar species and that the Australian specimen is not conspccific with those occurring further west and north in South-East Asia. We also found that a female conspccific with the Kakadu specimen was likely collected by Captain Phillip Parker King during his surveys of the northern Australian coast in about 1820 and described in 1826 by W. S. Macleay as Cenchroma obscura. King’s weevil has been forgotten for over 200 years, but the discovery of the Kakadu specimen suggests that this species, correctly named Hjpomeces obsfurus, may be present in northern Australia, albeit scarce and seemingly of no current agricultural concern. Introduction An unexpected discovery In September 2009 one of us (RGO) came across a ‘hairy’ yellow weevil (Figs 1, 2) in the insect collection of the CSIRO Tropical Ecosystems Research Centre (TERC) in Darwin. He recognised it as a species of Hypomeces Schoenherr, a genus distributed throughout South-East Asia, from eastern India and southern China southwards through Indochina and Indonesia to Timor and New Guinea, but not known to occur in Australia. Hypomeces currently comprises about ten species and belongs in the tribe Hypomeces weevils in Australia Northern Territory Naturalist (2016) 27 107 Tanymecini of the subfamily Entiminae, a large group of typically short-snouted weevils with wide host ranges as adults and soil-dwelling, root-feeding larvae. One species of Hypomeces, named H. squamosus (Fabricius) in the literature and ‘Gold- dust Weevil’ in vernacular language, is a major agricultural and horticultural pest in South-East Asia. Significandy, the specimen in the TF.RC collection (Figs 1, 2) is labelled as having been collected at the Naramu Camp of the former Kapalga Research Station in Kakadu National Park, Northern Territory, in April 1995 by Lyn Lowe, who then participated in a fauna survey forming part of the Kapalga Fire Experiment (Orgeas & Andersen 2001; Andersen et at. 2003). Moreover, the specimen, a male, is in a teneral condition (freshly cclosed), both its mandibles still carrying the deciduous cusp that occurs in Entiminae upon cclosion from the pupal case but breaks off when the weevil starts feeding, and its coating of yellow wax, which grows as the specimen ages and is more prominent in males, is only slighdy developed. Its teneral condition and pristine state of preservation indicate that the specimen was collected on the day it hatched from its cocoon and was pinned shortly afterwards, not stored in ethanol as this fluid would have dissolved its covering of wax and matted down its erect silvery setae. Comparison of the Kapalga weevil with specimens of I lypomeces in the Australian National Insect Collection (ANIC) in Canberra revealed that, although similar to the well-known Hypomeces ‘squamosuf (an invalid name, see below), it differs in a number of characters from this species and agrees more closely with specimens from Timor. The status of the Timorese taxon is unclear from the literature; it is sometimes treated as a ‘variety’ of H. ‘ squamosus ’ but has also been named as a different species. In their recent catalogue of Australian weevils, Pullen et al. (2014) settled on calling it I lypomeces rusticus (Weber, 1801), following the distinction made between this and H. '’squamosus' by Marshall (1916) in his scholarly treatment of the weevil fauna of British India. However, Pullen et al. (2014) changed the name Marshall had used for it, / lypomeces unicolor (Weber, 1801), to H. rusticus , in accordance with a recent correction published by Ren et al. (2013) and necessary due to the fact that Weber’s original name Cttrcu/io unicolor is a junior primary homonym of the older name Curculio unicolor Herbst, 1795 and hence nomenclaturally Figs 1-3. I lypomeces obscurtts (Macleay, 1826), male, Kapalga Research Station, Kakadu National Park, Australia. 1. dorsal view; 2. lateral view, 3. label. 108 Northern Territory Naturalist (2016) 27 Oberprieler et al. unavailable. For the same reason, H. squamosus had to be renamed as Hypomeces pulviger (Herbst, 1795) (Ren et al. 2013), an unfortunate but unavoidable change of the name of a well-known pest species. The identification of the Timorese taxon as H. rusticus remained somewhat insecure, however, as Marshall (1916) had expressed some doubt about the distinction of this species from 1l. pulviger (as H. squamosus), considering the few differences he could find between them to be ambiguous in some cases. Also, there is no recent and proper taxonomic study of the genus Hypomeces to verify them. Due to the fresh nature of the Kapalga specimen, Pullen et al. (2014) treated H. rusticus as occurring in Australia. Aims and objectives In this paper we report the results of further research into the taxonomy and nomenclature of the Kapalga weevil and outline the apparent history of the species in Australia. Although additional study is required (and in preparation) to fully resolve its taxonomic affinities, we here atm to draw attention to the indicated occurrence of this weevil in the Northern Territory and to list and illustrate the morphological differences between it and the more northerly pest species Hypomeces pulviger. We hope that this report will assist in the determination of whether this weevil species is established in northern Australia. Material and Methods We undertook a morphological study of 113 relevant specimens (including 13 types) of Hypomeces from the following collections: • ANIC — Australian National Insect Collection, Canberra, Australia; • MAGNT — Museum and Art Gallery of the Northern Territory, Darwin, Australia; • MMUS — Macleay Museum, University of Sydney, Sydney, Australia; • NAQS — Northern Australia Quarantine Strategy Entomology Collection, Darwin, Australia; • NHMD — Natural History Museum of Denmark, Copenhagen, Denmark. Selected specimens were photographed using a Jxtica DFC500 digital camera mounted on a Eeica M205C microscope, combining (“montaging”) image stacks in Ixtica Application Suite 4.4 and cleaning and enhancing the final images as necessary in Adobe Photoshop CS3. The genitalia of 15 specimens (mostly males) from different localities were dissected in the standard manner, temporarily stored in glycerine or KY Jelly® and photographed using the same equipment. Results Captain King’s lost weevil No other Australian specimen of Hypomeces has been located in any collection so far, but Zimmerman (1993: 667), in his bibliographic notes on William Sharp Macleay, Hypomeces weevils in Australia Northern Territoty Naturalist (2016) 27 109 <^<5* \ 1 ly P omeces ohs ™™s (Macleay, 1826) and H. pu/tiger (Hcrbst, 1795), dorsal aspect o head and prothorax. 13 . / 1. obscurus , male, Kapalga Research Station, Kakadu National ar ' Australia; 14. I I. obscurus , female, Pantc Macassar, Oc-Cussc, Timor-Leste; 15. / I. pulviggr, m.i c, atham, Laos; 16. II. obscurus , female, Chatthin Wildlife Sanctuary, Myanmar, (s — scape, mrg — median rostral groove, ard - admedian rostral depression, rp - tooth-like projection, mpg - median pronotal groove, apd - admedian pronotal depression, tpd - transverse pronotal Hypomeces weevils in Australia Northern Territory Naturalist (2016) 27 113 Figs 17, 18. I [ypomeces obscurus (Macleay, 1826) and H. pulviger (Herbst, 1795), lateral aspect of head and prothorax. 17. / I. obscurus , male, Kapalga Research Station, Kakadu National Park, Australia; 18. H. pulviger , male, Kuala Lumpur, Malaysia. Figs 19-22 (at left). I [ypomeces obscurus (Macleay, 1826) and / 1. pulviger (Herbst, 1795), aedeagus, lateral view. 19. I1. obscurus , Kapalga Research Station, Kakadu National Park, Australia; 20. 11. obscurus , Pante Macassar, Oc-Cusse, Timor-Leste; 21. / 1. pulviger. , Kuala Lumpur, Malaysia; 22. H. pulviger, India. 114 Northern Terri tor)' Naturalist (2016) 27 Oberpricler et al. What then is the correct name for the Timorese and Australian specimens? The oldest species name in contention is rusticus , which was given by Weber (1801), and also by Fabricius (1801), to specimens collected by the Danish naturalist O. K. Daldorff in Sumatra, probably at Bengkulu (Reid & Beatson 2015). Photos of the two type specimens of rusticus in Fabricius’ collection, kindly provided to us by the Natural History Museum of Denmark in Copenhagen, show these to possess a strong prothoracic tooth and thus not to be conspccific with the Australian and Timorese specimens (but apparently representing H. pulviger). The next oldest name is lanuginosa, which was proposed by Dejean (1821) for a species in Timor but not accompanied by a description and which is therefore unavailable for nomenclatural purposes (it was also never validated afterwards). Next in line of nomenclatural priority is obscura , which was established by Macleay (1826) with a proper description and is therefore nomcnclaturally available, although it has not been used for almost two centuries. Given the existence of the holotype of obscura in the Macleay Museum and its agreement in characters with the Kapalga and Timorese specimens (rather than with Hypomecespulviger), this species is to be named Hypomeces obscurus (Macleay, 1826) — the ending of the adjectival species name changing to accord with the different gender of the genus name (Hypomeces is masculine, Cenchroma feminine). Table 1. Differences between Hypomeces obscurus (Macleay, 1826) and Hypomeces pulviger (Herbst, 1795) (see Figs 9-22). Structure Hypomeces obscurus (previously 1 /. rusticus) Hypomeces pulviger (previously / 1. squamosus) Body scales colour always creamy; separate from each other colour usually iridescent green, at least in male; partly overlapping Rostrum longer shorter Admedian linear depressions on rostrum indistinct, very shallow, straight distinct, deep, curved Antennal scapes longer shorter Eyes flatter, less prominent more acute, very prominent Anterolateral comers of prothorax never tooth-like extended usually tooth-like extended Median pronotal groove shallow, indistinct deep, distinct (sharply edged) Pronotal impressions broad, shallow, transverse impression across base of median groove pair of short, narrower, deeper, irregular longitudinal impressions parallel to median groove Elytral bases more strongly rounded less rounded, partly straight Elytral setae of female verv fine, slightly longer shorter and thicker Penis shorter, more strongly curved; dorsally more open longer, less curved; membranous dorsal strip narrower Hypomeces weevils in Australia Northern Territory Naturalist (2016) 27 115 Discussion Captain King’s voyages and collecting localities Having clarified the identity and taxonomic status of Captain King’s weevil, its origin remains to be determined. Captain Phillip Parker King (1791-1856) was one of the famous Australian explorers of the 19th century. He undertook four voyages around Australia between 1817 and 1822, charged by the British Admiralty' and the Colonial Office to survey the north-west coast of New Holland, which his predecessor, Matthew Flinders, had not been able to chart during his circumnavigation of Australia in 1802- 1803. The Admiralty' thus instructed King to “examine the hitherto unexplored Coasts of |the Continent of] New South Wales, from Arnhem Bay, near the western entrance of the Gulf of Carpentaria, westward and southward as far as the North-West Cape,...”, and specifically to discover “any river or that part of the coast likely to lead to an interior navigation into this great continent.”. The Colonial Office wanted him “to obtain information” of, i.a., the “general climate ...”, the “directions of the mountains ...”, the “animals, whether birds, beasts, or fishes; insects, reptiles, &c., ...”, the “vegetables ... applicable to any useful purposes, ...” and the “descriptions and characteristic differences of the several tribes or people on the coast” (King 1827). On his first voyage, from December 1817 to July 1818, King sailed his sole ship, the cutter Mermaid , around the south and west coast of Australia and got as far east, on 26 March 1818, as Braithwaitc Point on the coast of western Arnhem Land. He then turned westwards again, exploring the nearby Goulburn Islands and surveying the coasts of the Cobourg Peninsula, Van Diemens Gulf and Melville Island before heading to Timor to reprovision his ship and then returning to Sydney. On his second voyage, from May 1819 to January 1820, he sailed the Mermaid northwards along the Australian east coast, around Cape York and across the Gulf of Carpentaria and explored the Arnhem Land coast from the Wessel Islands to Bathurst Island as well as the Cambridge and Admiralty Gulfs on the Kimberley coast, then ran for Timor again to take on provisions and home to Sydney along the west coast. On his third voyage, from June 1820 to December 1820, he followed the same route, but the Mermaid was “nail-sick” (leaking badly) by then and allowed him little opportunity for exploration, and he limped back from the Prince Regent River mouth to Sydney, this time without replenishing in Timor. On his fourth voyage, from May 1821 to April 1822, he had a new and larger ship, the brig Bathurst, which he again sailed around Cape York and the Gulf of Carpentaria to the Goulburn Islands, but he surveyed and explored mainly the coast of the western Kimberley region south to the Dampier Peninsula, returning to Sydney via Mauritius. Although King failed to find the fabled waterway into the interior of Australia, he explored practically every inlet along the north-western coast of Australia for about 1200 km west of Cape Wessel. King published a two-volume Narrative of his surveys soon afterwards (King, 1827), and a comprehensive and splendid account of his voyages, as well as of the many trials and tribulations he and his crew experienced during them, was published by Hordern (1997). 116 Northern Territory Naturalist (2016) 27 (Ibcrprielcr ct a / Captain King was given two marine surveyors, Frederick Bedwell (1796-1853) and John Septimus Roc (1797—1878), to assist him in this task, and the botanist Allan Cunningham (1791-1839) joined him in Sydney. The animals collected by King, Cunningham and Ro e on these voyages were studied and described in Appendix B of Volume II of King’s Narrative. William Sharp Macleay (1792—1865) studied the ‘Annulosa’, the ringed or segmented animals, the majority' (188) being insects, among them 108 beetles (Colcoptcra) and among these 20 weevils (Curculionidae) (Macleay 1826). He described nine of the weevils as new, although some turned out to have already been described by earlier authors and others belonged to different genera than those to which Macleay assigned them (Zimmerman 1993). Macleay did not provide the names of the collectors of these beedes or the localities where they were taken, and not all occur along the ‘intertropical and western coasts’ of Australia. The weevil specimens he named Cenchroma lanuginosa evidendy originated from Timor, not only because this name had been published by Dejean (1821) for a species from Timor (and Dejean was an acquaintance of Macleay) but also because the two specimens with this name in the Macleay Museum carry a label reading “Timor”. King briefly visited the harbour of Kupang in western Timor on his first two voyages, and Cunningham collected specimens (mainly plants) in the vicinity of the town on both occasions (Hordern 1997; Orchard & Orchard 2013). In contrast, the single specimen of Macleay’s Cenchroma obsatra is labelled as “Australasia”, in Macleay’s hand, suggesting that it was not collected together with the two males from Timor but separately and from somewhere else. But where? Looking for a weevil in a haystack The name “Australasia” was coined in the 18th century for the lands south of Asia, so encompassing Australia, New Zealand, New Guinea and their neighbouring islands in the Pacific Ocean (but not Timor). In King’s and Macleay’s times, the name “Australia” was not yet established and commonly used for the Australian continent, which was generally referred to as “New Holland” or, as on King’s instructions from the British Admiralty, “New South Wales”. Macleay used both names “Australasia” and “New Holland” on the labels of his insect specimens, the former probably when he was unsure of their exact origin. The holotype of Cenchroma obscura is not the only Macleay type labelled as having come from “Australasia”. Among the Macleay types in the ANIC there are another 11 with the same locality name on the label (Acanthocinus piliger ; Callidium erosum, Chrysomela k/ttgii, Chrysomela nigrovaria , Cistela securifera, Clems cntciatus, Coccinella king, Lyots septemcavus, Lycits rhipidittm, Notoclea sp/endens, Te/ephonts pulchellus), whereas nine others (Chryso/optts echidna, Chrysolopus tuberculatus, Tllater nigroterminatus, FJater xanthomma, Epholosium velutinum, Hybanchenia nodulosa, Oedemerapunctum, Talaurinus kirhyt, Trox alter nans) are labelled as from “New Holland” instead. Most of the species whose Macleay types are labelled “Australasia” do not occur in Timor but only in Australia, i.c. the coccinellid Coccinella kingi (now' Archegleis kingi; Pope 1989; Slipiriski 2007; Adam Slipinski, pers. comm.), the cantharid Telephones pulchellus (now Cbauliognathus luguhris (Fabricius)), the lycids Lyctts rhipidittm and L. septemcavus (now both Porrostoma rhipidium ; Lodislav Bocak, Hypomeces weevils in Australia Northern Territory Naturalist (2016) 27 117 pers. comm.) and the cerambycids Acanthocinus piliger (now Rbytidophora piligera\ Adam Slipinski, pers. comm.) and CalUdium erosum (now Pytbeus erostts; Adam Slipinski, pers. comm.). The exact distribution ranges of the species described by Macleay from King’s material are often not known; some of them are widespread in Australia and others are restricted to the south-eastern or south-western parts, but at least two occur in the Northern Territory and northern Western Australia, i.e. the clerid Ckrus cntciatus (now Orthrius cntciatus) and the tenebrionid Cistela seettrigera (now Nocar securigents). It is thus manifest that most of the beedes described by Macleay (1826) and labelled “Australasia” must have been collected in Australia, and some indeed likely in the Northern Territory or north-western Western Australia, and there is no primafacie evidence that the type of Cenchroma obscura was not collected there either. King and his crew explored almost the entire north-west coast of Australia and went ashore on many islands and points and bays on the mainland, and especially Cunningham collected specimens wherever and whenever he could (Curry el al. 2002). Among the likely places he (or King or Roe) could have taken the type of Cenchroma obsettra are South Goulburn Island and Sims Island, where Cunningham collected specimens on all four of King’s voyages, and especially the banks of the South Alligator River, which King and Cunningham explored upstream for about 64 km from its mouth on the first voyage and where, on 8 May 1818, they collected near the present site of Kapalga (Curry et al. 2002: Map 8). In his journal Cunningham recorded some plants he encountered there on that day but nothing about any insects, but as his journal entries generally only deal with botanical specimens (Tony Orchard, pers. comm.), this does not mean that he could not have taken such a weevil there. An exact locality' for the type of Cenchroma obscura can probably never be established, but it is very likely that it was indeed collected along the Northern Territory coast. No further specimens of H. obscurus have been found in Australia to date, despite 25 years of quarantine inspection of numerous locations in the Northern Territory by the NAQS team in Darwin (Glenn Beilis, pers. comm.). A recent search at the Kapalga site also failed to find another specimen, but it was undertaken in July 2015, in the dry season when the parched condition of the vegetation gready reduces insect activity. The absence of further specimens so far suggests that, if the species is present in the Northern Territory, it may have a restricted distribution and/or occur in very low numbers, and the time of collection of the Kapalga specimen (April) and also of King’s 1818 visit to the site (March) indicate that it may only be active during the wet season. Potential impact The indicated occurrence of a Hypomeces species in northern .Australia is important as H. pulviger remains a target (under the name / /. sc/uamosus) of quarantine surveillance efforts in the area (Glenn Beilis & Luke Hailing, pers. comm. 2015). This notorious pest (the ‘Gold-dust Weevil’) has a wide range of hosts in South-East Asia. I lill & Abang (2006) recorded it from 42 hosts in Malaysia alone. The highly polyphagous nature of both 118 Northern Territory Naturalist (2016) 27 ( )bcrprielcr et al adults and larvae can cause significant damage on a number of agricultural crops, the major hosts being rice, maize, sugarcane, cotton and tobacco (Kalshoven, 1981), along with Citrus spp. and sweet potato (Hill, 2008). Other hosts include cocoa, coffee, durian, guava, jackfruit, long-bean, mango, rambutan and sapote (Muniappan et al. 2012), and additional ones are listed, together with a summary of the weevil’s impact on crops and additional references, on CABI’s Plantwise Knowledge Bank (http://www.plantwi.se. org/KnowledgeBank/Datasheet.aspx?dsid=27783). In contrast, little information exists about the hosts of H. obscurus in Timor. Specimens in the AN1C have been collected on Pigeon Pea ( Cajanus cqjart, Fabaceac) and Jujube or Chinese Apple ( Zispphus mauritiana, Rhamnaceae) in West Timor, and it has been found defoliating mango and was also taken on guava, maize, long-bean, peanut, sweet potato, sorghum, cucumber and rice (Glenn Beilis, pers. comm.). This host range suggests that H. obscurus may also be able to feed on a variety of plants (both native and cultivated) in Australia if it is established here now or in the future. Conclusions Our intricate sleuthing work revealed that Lyn Lowe, quite unbeknown to her, succeeded in rediscovering Captain King’s lost weevil in the Kakadu National Park and diat the name William Sharp Maclcay gave it, forgotten in the scientific literature for almost 200 years, is in fact valid. While it seems impossible to determine the exact locality where King and his party may have collected this specimen nearly 200 years ago, King and Iris botanist, Allan Cunningham, did collect specimens in the vicinity of Kapalga, the site where Lyn Lowe took a freshly hatched male in 1995. As far as currently known, Hypomeces obscurus occurs mainly on Timor, and it is not the same species as Hypomeces pulviger (formerly H. squamosal), the notorious “Gold-dust Weevil” (a misnomer as its colour is neither golden nor due to dust). Further collecting efforts at the Kapalga site as well as in similar habitats elsewhere in Kakadu National Park and other parts of die Northern Territory are needed to confirm the presence of H. obscurus in Australia and verify whether King’s lost weevil is indeed alive and well in the Northern Territory. Such confirmation would indicate that the species is either native to Australia or was transported there by humans (e.g. by Indonesian fishermen) at least two centuries ago and has been established for a considerable time. Acknowledgements We sincerely acknowledge a range of associates who helped us in this complex investigation to discover the status and name of King’s lost weevil: Alan Andersen (CSIRO, TERC, Darwin) and Lyn Lowe (Charles Darwin University', Daman) for validating the collecting details of the Kapalga specimen; Robert Blackburn and Jude Philp (MMUS) for the loan of King’s original specimens; Gavin Dally, Graham Brown and Richard Willan (MAGNT) and Glenn Beilis and Stacey Anderson (NAQS) for the loan of Hypomeces specimens from their collections; Miguel Alonso-Zarazaga (Museo Nacional de Ciencias Naturales, Madrid, Spain) for the names and locations of several other specimens; Sree Hypomeces weevils in Australia Northern Territory Naturalist (2016) 27 119 Selvantharan (NHMD) for the excellent photos of critical Weber and Fabrician types; Glenn Beilis, Stacey Anderson and Greg Chandler (NAQS) for a preliminary molecular analysis of a number of critical specimens and for various pieces of information on Hypomecer, Luke Hailing (Australian Quarantine and Inspection Service, Cairns) for information on Hypomeces pest species; Adam Slipinski (CSIRO, ANIC) and Ladislav Bocak (Olomouc, Czech Republic) for information on the distribution of beedc species described by Maclcay; Tony Orchard (Canberra) for information on Allan Cunningham’s records and journals; Ted Edwards and Russell Barrett (CSIRO NCRA) for important literature;John Wcstaway (NAQS) and Rachel Martin (Parks Australia, Kakadu) for help in searching for further specimens at the Kapalga site. References Andersen A.N., Cook G.D. and Williams R.J. (2003) Fire in Tropica!Savannas. The Kapalga Experiment. Springer Science & Business Media, New York. Curry S., Maslin B.R. and Maslin J.A. (2002) Allan Cunningham Australian Collecting localities. Australian Biological Resources Study, Canberra. Dejean P.F.M.A. (1821) Catalogue de la collection de Coleopteres de M. k Baron Dejean. Chez Crevot, Libraire, Paris, i-viii, 1—138 pp. Fabricius J.-C. (1801) Systema ekutheratorum secundum ordines, genera, species: adiectis synonymis, locis, observationihus, descriptionibus. Tomus II. Bibliopolii Academici Novi, Kiliae. Hill D.S. (2008) Pests oj Crops in Warmer Climates and Their Control. Springer Netherlands. Hill D.S. and Abang F. (2006) The Insects of Borneo (Including South-East and East Asia). Univcrsiti Malaysia Sarawak, Sarawak. Hordern M. (1997) King of the Australian Coast. The Work of Phillip Parker King in the Mermaid and Bathurst 1817—1822. (Paperback E.dition, 2002). Melbourne University Press, Carlton. Kalshoven L.G.E. (1981) Pests of Crops in Indonesia. Revised and Translated try P.A. van der Laan; with the Assistance of G.H.L. Rothschild. Ichtiar Baru, Jakarta. King P.P. (1827) Narrative of a Survey of the Inlertropical and Western Coasts of Australia, Performed between the Years 1818 and 1822. I bis. I, II. John Murray, London, 451 + 637 pp. Macleay W.S. (1826) Annulosa. Catalogue of Insects, collected by Captain King, R. N. Appendix B. Containing a list and description of the subjects of natural history collected during Captain King’s survey of the intertropical and western coasts of Australia. In: Narrative of a Surrey of the Intertropical and Western Coasts of Australia, Performed between the Years 1818 and 1822. 1 oi II. (ed. King P.P.), pp. 438-469, pi. B. John Murray, London. Marshall G.A.K. (1916) Coleoptera. Rhynchophora: Curculionidae. In: The Fauna of British India, including C.eylon and Burma (ed. Shipley A.E.). Taylor & Francis, London, xv + 367 pp. Muniappan K., Shepard B.M., Carner G.R. and Ooi P.A.C. Arthropod Pests of Horticultural Crops in Tropica!Asia. CAB1, Wallingford, UK. Orchard A.E. and Orchard T.A. (2013) Allan Cunningham’s Timor collections. Nuy/sia 23, 63—88. Orgeas J. and Andersen A.N. (2001) Fire and biodiversity: responses of grass-layer beedcs to experimental fire regimes in an Australian tropical savanna. Journal of Applied Ecology 49,62. Pope R.D. (1989) A revision of the Australian Coccinellidae (Coleoptera). Part 1. Subfamily Coccinellinae. Invertebrate Taxonomy 2, 633—735. Pullen KR., Jennings D. and Obcrprielcr R.G. (2014) Annotated catalogue of Australian weevils (Coleoptera: Curculionoidea). Zootaxa 3896, 1-481. 120 Northern Territory Naturalist (2016) 27 Oberprieler et al. Reid C.A.M. and Beatson M. (2015) Disentangling a taxonomic nightmare: a revision of the Australian, Indomalayan and Pacific species of Attica Geoffrey, 1762 (Coleoptera- Chrysomelidae: Galerucinae). Zootaxa 3918 (4), 503-551. Ren L., Sanchez-Ruiz M. and Alonso-Zarazaga M.A. (2013) Family Curculionidae I .atrcillc, 1802- subfamily Entuninae Schocnhcrr, 1923: tribe Tanymecini Lacordaire, 1863. In: Catalogue of Pataearetic Coleoptera. Volume 8. Curculiomidea II (eds Lobl I. and Smetana A.), pp. 392-413. Brifi Leiden. Slipirriki A. (2007) Australian ladybird Hectics (Coleoptera: Cocdnellidae). Their Biology and Classification. Australian Biological Resources Study, Canberra. Weber F.l. (1801) Observationes entomological, continentes novorum e/nae condiditgenerum characteres, et nt/per detectarum specierum descriptiones. Bibliopolii Acadcmici Nova, Kiliae, xii + 116 pp. Zimmerman JG.C. (1993) Australian Weevils (Coleoptera: Curculionoidea). Volume III. Nanop/yidae, Rhyncbopboridae, Urirbinidae, Curculionidae: Amycterinae, literature Consulted. CSIRO Australia Melbourne. Northern Territory Naturalist (2016) 27: 121-125 Short Note Rediscovery of the Spinifex Sand-skipper (Proeidosa polysema) in the Darwin area, Northern Territory Michael F. Braby 1 ' 2 and John O. Westaway 3 1 Division of Evolution, Ecology and Genetics, Research School of Biology, The Australian National University, Canberra, ACT 0200, Australia Email: michael.braby@anu.edu.au 2 The Australian National Insect Collection, GPO Box 1700, Canberra, ACT 2601, Australia ’Northern Australia Quarantine Strategy, Commonwealth Department of Agriculture, 1 Pederson Road, Marrara NT 0812, Australia Abstract The Spinifex Sand-skipper (Proeidosa polysema) (Lepidoptera: Hesperiidae) is recorded from two sites near Noonamah-Berry Springs, approximately 28 km south-east of Darwin. The species is recorded breeding on the grass Triodia bitextura (Poaceae) growing in cucalypt open-woodland in sandy soil derived from laterite. The butterfly had not been recorded from the Darwin area for more than a century (since 1909) and its presence in the rural area confirms earlier historical collections made by renowned entomologist F.P. Dodd. Introduction The Spinifex Sand-skipper (Proeidosa polysema) (Fig. 1), has a wide distribution across the northern half of Australia where it occurs in hummock open-grassland on sand dunes and cucalypt open-woodland with a hummock/tussock grass understorey on sand and dry rocky sandstone, preferring shallow gullies and slopes of hills (Braby 2016). The larvae (Figs 3-5) specialise on a limited range of perennial ‘soft’ resinous spinifex tussock-forming grasses (Triodia spp.) (Poaceae), which in the Top End of the Northern Territory include T. microstachya (Common & Waterhouse 1981) and T. bitextura (Braby 2015). The larvae construct distinctive tubular shelters by joining several blades of grass together with silk (Fig. 2); the entrance to the shelter is located at the bottom and the larva typically rests upside-down during the day, emerging at night to feed on the resinous leaf blades. During die dry season, especially in more arid areas when food quality declines, the larva docs not feed and remain in diapause (I'ig. 5) for many months. The larva also pupates in the final instar larval shelter. In the Top End, the northern-most occurrences of P. polysema are Kakadu National Park (Nourlangie Rock) and the Robin Falls area, including the sandstone escarpment of the falls proper (MFB, pers. obs.) and the country behind the radio repeater station approximately 2 km north-west of the falls to the south of the Adelaide River township 122 Northern Territory Naturalist (2016) 27 Braby & Westaway Figs 1-5. Proeidosapofysema showing: 1. adult male hilltopping on sand dune in central Australia at Curtin Springs, Northern Territory; 2. larval shelter on Triodia hitextura at Wongalara Wildlife Station, Northern Territory; 3. early instar larva on T. microstachya at Fish River Station, Northern Territory; 4. final instar larva on T. microstadrya at Fish River Station, Northern Territory; 5. final instar larva in diapause on T. pungent at Curtin Springs, Northern Territory. (Michael Braby) (Meyer et al. 2006). Historical records further north in the Darwin area (‘Port Darwin’) have been dismissed or considered to require confirmation because of the lack of subsequent records for over a century. Indeed, Meyer et al. (2006: 15) stated that: “We believe it is unlikely that it will be encountered there in the future, due to a lack of suitable habitat.” Waterhouse (1933) and Meyer et al. (2006) reviewed the historical literature records of P. po/ysema from the Darwin area and noted that at least five specimens were collected from Port Darwin by F.P. Dodd at the turn of the twentieth century: two males (both paratypes) in February 1909, other males in January and March, and a female in April. It has generally been assumed that the material was collected further south, possibly along the railway line such as between Adelaide River and Pine Creek, and that Port Darwin was the location at which the specimens were processed. In this note, we record an extant population of P. polysema from the outer Darwin rural area. Prior to our rediscovery of Spinifex Sand-skipper rediscovered Northern Territory Naturalist (2016) 27 123 the species near Darwin, the butterfly had not been recorded from the area for more than a century. Observations At the intersection of Finn Road and Middle Arm Road, approximately 6.5 km north of Berry Springs (12.6451°S, 131.0099°E), an adult of Proeidosapolysema in ‘fresh’ condition Figs 6-11. Ecology of Proeidosa polysema near Berry Springs-Noonamah, Darwin: 6. breeding habitat comprising eucalypt open-woodland with a grassy understorey dominated by 7 nodi a bilexlurir, 7. larval food plant Triodia bitextura\ 8-11. adult butterfly being eaten by preying mantis. (Michael Braby) 124 Northern Territory Naturalist (2016) 27 Braby & Westaway (according to extent of wing wear) was observed just after midday feeding on the flowers of Spermacoce sp. on 14 February 2015. The specimen was observed only briefly before it was disturbed; it rapidly flew off and could not be relocated despite extensive searching. However, another adult was subsequently located nearby and photographed; however, this particular individual was being eaten by a preying mantis (Figs 8-11). After this insect predator had completed devouring its meal, two of the wings ot the butterfly were recovered and retained as vouchers (accession number MFBC 00934, Australian National Insect Collection). The identity of the butterfly can be clearly discerned by the uniform brown ground colour to the wings and the presence ot a series of eight large white spots on the underside of the hindwing (Fig. 8). The habitat in which these observations were made comprised an open disturbed area of pioneer plants adjacent to the railway line. There were no signs of the putative larval food plant {Triodia bitextura) growing in the immediate vicinity. Two months after these initial field observations, we searched the adjacent woodland for presence of the larval food plant and likely breeding areas of the butterfly on 26 April 2015. On Middle Arm Rd, approximately 6 km north-north-east of Berry Springs and 6.4 km west-south-west of Noonamah (12.6456°S, 131.0233“E), we found Triodia bitextura (voucher JOW 4802, Darwin Northern Australia Herbarium) growing in abundance in open-woodland with a dense grassy understorey in sandy soil derived from well-drained laterite on relatively flat terrain (Fig. 6). On one particular tussock (Fig. 7), two old pupal shelters of P. polysema were recorded and collected, confirming the presence of an extant breeding site. This site was located approximately 1.5 km cast of the site where adults were initially recorded earlier in February 2015. The breeding habitat was located approximately 28 km south¬ east of Darwin. Discussion Our discovery of Proeidosa polysema from the Darwin area near Noonamah and Berry Springs validates the historical records from ‘Port Darwin’ by Dodd as being reliable. Moreover, a distribution map of the larval food plant Triodiabitextura for the Darwin area (Fig. 12) shows that this plant is widespread but uncommon in the Darwin rural area, extending as far the Darwin area, together with the two extant sites of Proeidosa polysema (★). Symbols for plant data are as follows: • vouchered herbarium specimens, A survey observations. (Flora and Fauna Division) Spinifex Sand-skipper rediscovered Northern Territory Naturalist (2016) 27 125 north as Gunn Point. In our experience, T. bitextura has a scattered occurrence and in some areas it may be locally abundant. Trioclia bitextura (Fig. 7) does not exhibit the large hummock forming habit typical of ‘hard’ spinifex species, and as such it is less conspicuous. Moreover, it is not confined to sandstone outcrops, which may explain the belief of Meyer et ai (2006) that the butterfly does not occur in the Darwin area due to lack of suitable habitat. At the Noonamah-Berry Springs site the butterfly was breeding on plants growing on sandy soil derived from lateritc in contrast to the general tendency of P. polysema to breed on sand dunes and sand derived from rocky sandstone. The spatial distribution of T. bitextura suggests the butterfly may well occur in other locations near Darwin, particularly in the rural area to the south and south-east of Palmerston (Fig. 12). However, little habitat now remains closer to the city of Darwin, but it seems highly plausible that in 1909 Dodd collected the original specimens not far from his base camp in Parap (see also Braby & Nielsen 2011 for an account of Dodd’s collecting sites in the Darwin region), rather than along the railway line to the south of Adelaide River. References Braby M.F. (2015) New larval food plant associations for some butterflies and diurnal moths (Lepidoptera) from the Northern Territory and Kimberley, Australia. Part II. Records of the Western Australian Museum 30, 73-97. Braby M.F. (2016) The Complete Field Guide to Butterflies of Australia. Second Edition. CSIRO Publishing, Melbourne. Braby M.F. and Nielsen J. (2011) Review of the conservation status of the Atlas Moth, Attacus wardi Rothschild, 1910 (Lepidoptera: Saturniidae) from Australia, journal of Insect Conservation 15, 603-608. Common l.F.B. and Waterhouse D.F. (1981) Butterflies of . Australia. Angus and Robertson, Sydney. Meyer C.E., Weir R.P. and Wilson D.N. (2006) Butterfly (Lepidoptera) records from the Darwin region. Northern Territory. The Australian Entomologist 33, 9-22. Waterhouse G.A. (1933) Notes on the type specimens of Hesperiidae (Lepidoptera) in the museums in Australia, with special reference to those in die South Australian Museum. Records of the South Australian Museum 5, 49-62. Northern Territory Naturalist (2016) 27: 126—128 Book Review Book Review John Dengate Avalon, NSW 2107, Australia When 1 was asked to review Dr Cirahamc Webb’s (2015) book Wildlife Conservation: In the Belly of the Beast, 1 was a bit uneasy. Cecil, the tame old lion at Hwange National Park in Matabelcland North, Zimbabwe, had just met an untimely end and 1 had heard Dr Webb’s name associated with trophy hunting. But having come to grips with the book, I must say it’s a valuable contribution to conservation and the issues that surround it. Dr Webb writes that if his book “stimulates others to think in more depth about conservation, or helps them better understand and appreciate how bio-politics can enhance or constrain conservation” then his main goal in writing the book will be achieved. Well I think he can rest assured on that point - the book contains a plethora of examples of rational science being ignored in favour of ‘bio-politics — the rarely-mentioned wheeling and dealing that goes hand in hand with a lot of conservation decisions. For instance, the ‘sea turtle conservation community’ estimated the total Caribbean Hawksbill Turtle population as 5000, even though Cuba harvested 5000 adult turtles every year. Armed with reasonable science showing the harvest was sustainable, Cuba applied to CITES (die Convention on International Trade in Endangered Species) to export 500 turde shells to |apan. It was blocked by the United States of America and conservation interests. Dr Webb makes a compelling case that the United States was motivated by political factors like blockading Cuba and getting the votes of expatriate Cubans living in the United States. Of more concern is the conclusion that conservation groups were desperate to continue using the Hawksbill as one of their iconic fundraising species - something that would be less than convincing if it were admitted that Hawksbills were thriving in Cuba as part of a sustainable use program. You might argue that legal trade in secure Cuban Hawksbills would be the best cover for illegal trade in endangered populations from elsewhere, but prohibition on trade in wildlife hasn’t exaedy been successful - both Tigers and Black Rhinoceros have come much closer to extinction after international trade was banned. Cuba eventually caved in to the pressure and banned the 1 lawksbill harvest - bringing a 500 year old tradition to an end. The World Wildlife Fund is providing the turtle fishing communities with “sustainable economic alternatives” (http://www.wwf.ca/about_us/ successes/hawksbill/). 1 can’t help wondering if they will support the communities in the long term. Book Review Northern Territoty Naturalist (2016) 27 127 One of the most iconic conservation issues is that of elephants, and the ivory they produce. Dr Webb points out that “given the opportunity', wild elephants will continually multiply until they ultimately destroy the habitat in which they live.” I’ve seen a graphic demonstration of this in Kenya’s Amboseli National Park where a few remaining monkeys clung pathetically to tree stumps - the last remnants of a woodland eaten out by elephants. Dr Webb argues that instant death from culling is far more humane than a slow, agonising end from starvation. He further points out that some of the culling can be done by controlled trophy hunting, with the resulting revenue benefiting local communities - who then value the animals as a source of income, rather than seeing them as an agricultural pest. This may be a distasteful argument, but it also seems to be a compelling one - provided the controls on the program ensure a humane and sustainable harvest. A similar point can be made with trophy hunting of crocodiles in the Northern Territory. Dr Webb’s personal involvement in the commercial aspects of crocodiles gives him a valuable perspective on the sometimes perverse results of the costs of wildlife regulation. I guess the government position is that any industry using a protected species should fund the regulatory regime that ensures the species is harvested sustainably and humanely. But in the case of crocodiles, the costs of CITES compliance could be having a perverse impact on the species. For example, a crocodile has 66 teeth, each of which you can sell for between $5-$10, once you have cleaned, drilled and mounted it on a leather thong. But to sell that crocodile’s teeth overseas, each tooth needs a separate export permit from Australia and in some cases, an import permit from die destination. Each permit takes about 40 working days and about $60 in fees and costs. So to sell the teeth from one farmed crocodile, could take nearly $8000 in permit fees. OK, the skin is the main product, but one impact of this cost structure is to favour agriculture over wildlife production — with consequent damage to natural ecosystems. Another likely impact of this cost structure is to drive the trade underground - if it’s impossible to comply with the costs of CITES, then people will look for underground ways to sell their produce - especially in hard-pressed third world countries. I can’t help thinking that at the heart of this book is the clash of the generations. On the one hand, Dr Webb provides compelling rational arguments for the controlled commercial use of wildlife, but you can hear his (mostly younger, urban) opponents saying how they hate the idea of killing anything and especially dislike firearms. Perhaps this difference stems from today’s society being separated from the realities of despatching farm animals for food. I suspect that our farming grandfathers would find 128 Dengate Northern Territory Naturalist (2016) 27 current attitudes to wildlife rather inconsistent — at least for those of us who cat meat or fish. And speaking of attitudes, one of the inescapable lessons from this book is that unfortunately, scientific facts frequently come a poor second to emotion when it comes to why people adhere to particular beliefs. Emotional responses often come from experiences. So politicians probably won t have any personal support for a cause unless they have had a relevant experience. And even when they have, they won’t take action unless it is OK with the electorate. Which is why using the media to get grass roots support for something is so important. And also why the most prominent conservation and animal welfare organisations are the most expert at using the media. In this controversial book, Dr Webb applies scientific method to the field of conservation and often finds it wanting. The book is full of examples of conservation decisions taken for political reasons against scientific evidence — it’s a thought-provoking work — and a scary read! Everyone should have a look at it. Reference Webb G. (2015) Wildlife Conservation: In the belly of the Beast. Charles Darwin University Press, Darwin, NT. Advice to authors The Northern Territory Naturalist publishes works concerning any aspect of the natural history and ecology of the Northern Territory or adjacent areas of northern Australia. It is a registered, peer-reviewed journal (ISSN 0155-4093) for original research. Contributors include a range of field naturalists and scientists who do not have to be members of the Northern Territory Field Naturalists’ Club Inc. Submission of manuscripts Manuscripts arc considered on the understanding that the content has not been published, accepted or submitted for publication elsewhere and that all authors agree to its submission. All manuscripts are refereed. The editors reserve the right to require modification of manuscripts to eliminate ambiguity and repetition and improve communication. There are no page charges. The editors welcome suggestions for colour plates, which may be included at no charge at the Club’s discretion provided the material contributes substantially to the manuscript and enhances the journal. The Northern Territory Field Naturalists’ Club Inc. retains copyright to all published manuscripts, while granting an unrestricted licence for non-commercial use by authors. A PDF copy of the published manuscript will be provided to the primary author. Authors may submit articles in the form of Reviews, Research Articles, Short Notes, Species Profiles or book Reviews. Reviews (up to 10,000 words) generally emphasise the synthesis of existing (published) data or knowledge rather than presenting new primary data or results. The choice and number of subheadings is optional, bur an abstract must be included. A Research Article (up to approx. 5000 words though longer papers will be considered at the discretion of the editors) is a succinct scientific paper describing the findings of original research written in the traditional format, including abstract, introduction, methods, results, discussion, acknowledgements, references anti appendices. Short Notes (up to approx. 1500 words) will be considered where the contribution significantly increases current knowledge of natural history, such as describing new or unusual field observations, or summarising survey methods. Subheadings are optional, but an abstract is essential A Species Profile consists of a short overview of a particular animal or plant. An abstract is required bur other subheadings are optional. Figures (e.g. distribution map) and other illustrative material arc welcome. Book Reviews (up to c.1000 words) of natural history works (books and CDs) of major importance to the Northern Territory or northern Australia are encouraged. Reviews, Research Articles, Short Notes, Species Profiles and Book Reviews should be written to be intelligible to readers without specialist knowledge of the subject. Authors should consult the most recent edition of the journal as a guide to layout, including headings, table format, and references. For more detailed instructions on preparation of manuscripts please refer to the guidelines for authors on the Northern Territory Field Naturalists’ Club Inc. website