Northern Territory Naturalist Number 18 November 2005 ISSN 0155-4093 The Journal of the NT Field Naturalists' Club FIELD NATURALISTS CLUB NORTHERN TERRITORY FIELD NATURALISTS' CLUB Founded 1977 Officers for 2005/06 President Secretary Treasurer NT Naturalist Editors Don Franklin Lyn Reid Fiona Douglas Flelen Larson Steve Reynolds Fiona Douglas ISSN 0155-4093 © 2005 Northern Territory Field Naturalists' Club Inc. 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 the Seminar Room of the School of Australian Indigenous Knowledge Systems at the Charles Darwin University Casuarina Campus, Darwin, at 7:45 pm on the second Wednesday of each month. All members receive the monthly newletter 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: http://www.geocities.com/ntfieldnaturaljsts Front Cover: The terrestrial orchid Didymop/exis pattens is known from only six locations in the Northern Territory. This leafless saprophyte survives most of the year as an underground tuber, emerging early in the wet season. (Don Franklin) Rear Cover: The elusive Chameleon Dragon Cbelosania brunnea has been recorded from scat¬ tered locations in the Top End and Kimberley, with the major:tv of records from Kakadu National Park. (Alex Dudley) Northern Territory Naturalist (2005) 18: 1-7 New location records for some butterflies of the Top End and Kimberley regions Donald C. Franklin 1,2 , Bruce Michael 2 and Max Mace 2 1 School for Environmental Research, Charles Darwin University, Darwin, NT 0909. Email: don.franklin@cdu.edu.au 2 Northern Territory Field Naturalists Club Inc., PO Box 39565, Winnellie, NT 0821. Abstract We report 52 new locations in north-western Australia for 22 species of butterfly. Six records are of three species rarely reported in the region, the Fiery Jewel Hypochrysops ignita , the Two-spotted Line-blue Nacaduba biocellata and the Long-tailed Pea-blue IM/upides boeticus. Records of the Orange Ringlet Hypogsta adiante , Spotted Pea-blue Euchrysops cnejus and Small Dusky-blue Candalides erinus , species previously reported as occurring patchily in north-western Australia, are sufficient to suggest that they occur throughout the region. The ease with which new location records were obtained beyond the Darwin area suggests that much remains to be learned about the distribution of butterflies in the remoter parts of north-western Australia. Introduction For butterflies, generic Australia-wide distribution maps covering all species have been available for over three decades (Common & Waterhouse 1981, and earlier editions). Dunn and Dunn (1991) mapped over 90 000 specimen records continent¬ wide. More recently, Braby (2000) has presented updated generic distribution maps incorporating Dunn and Dunn’s data set and other records. Nevertheless, Braby noted that "In many cases, gaps in the ranges shown .... do not necessarily reflect natural disjunctions". Braby goes on to name the arid zone, but surprisingly not the monsoon tropics, as a primary area of deficient reporting. Knowledge of invertebrate distributions has traditionally been and largely remains based on collections, with the exception of a few distinctive species (e.g. Wilson et al. 2003). The value of observational records for an entire faunal group has been amply demonstrated for birds (Blakers et al. 1984, Barrett et al. 2003), both of which have spawned numerous subsidiary analyses (e.g. Franklin et al 2000, Griffioen & Clarke 2002). However, harnessing such records is dependent upon the ability to accurately identify species in the field, a process that for butterflies in Australia has been enhanced by the production of a handbook (Braby 2000) and more recently, a field guide (Braby 2004). Braby acknowledges the use of "recording" as well as collections 2 Northern Territory Naturalist 18 (2005) D. Franklin et at. in the preparation of his maps, and Puccetti (1991) provided a tangible example of the value of observation-based recording of butterflies. In this note, we present records of butterflies observed in the Top End and Kimberley region of monsoonal northern Australia which are c. 100 km or more beyond the ranges mapped by Braby (2000). Methods Butterfly records were obtained during formal surveys and as incidental sightings: (1) during brief visits to the Victoria River District (Northern Territory') and Kununurra area (Western Australia) by BM and MM in May and August 2003; (2) during a 3-day survey of the butterfly fauna of Elsey National Park and vicinity by DF, BM and others in May 2004; and (3) in brief explorations of the butterfly fauna incidental to other research, during a series of expeditions to remoter parts of north¬ western Australia, including sub-coastal areas of the Gulf of Carpentaria, northern Arnhemland, and central and north Kimberley, during the dry' season of 2004 (DF). All the species reported here are, with appropriate experience, readily identified as adults either in the field or in the hand, skills that we have developed intensively over the last three years during surveys in the vicinity' of Darwin. Records of lycaenids are based on either close examination of netted specimens, or in a few cases, close examination of individuals at rest in the field. Most of our nymphalid records, along with the sole papilionid record, are based on sightings of butterflies in flight or at rest. In all cases, we have given due consideration to possible alternate identifications based on species known to occur in north-western Australia (Braby 2000), and in most cases, we had prior familiarity with both the species reported and of similar species. Unfamiliar species and most lycaenids were checked in the field against the illustrations of Braby (2000), the identification process being enhanced by a field key developed by one of us (DF) based primarily on the descriptions, identification notes and illustrations in Braby (2000). The field guide of Braby (2004) was not available during the study period. We have excluded records where there might be a reasonable call for a confirmatory voucher specimen (e.g. many 1 lesperiidae). We have not attempted to identify individuals to subspecies level. With the possible exception of the Fiery Jewel Hypochrysops ignita, it seems unlikely that information about subspecies, were it available, would be of particular interest because the records presented here are nested within more general distributions that involve only a single subspecies. Results Fifty-two new locations involving 22 butterfly species are summarised in Table 1, including 41 records of 19 species in the Northern Territory' and 11 records of seven species in Western Australia. Two species observed in Western Australia, the Dusky New butterfly records Northern Territory Naturalist 18 (2005) 3 Knight Ypthima arc/ous and Chocolate Argus Junonia bedouin , were not reported from that state by Braby (2000), but Grund and Hunt (2001) reported both species from Kalumburu and considered the Dusky Knight to be "common" there. Most locations were in the order of 100-200 km beyond areas indicated by Braby (2000), with the greatest being a c. 600 1cm range extension west-south-westwards for the Chocolate Argus. The record of the Long-tailed Pea-blue luuupides boeticus at Gove Peninsula is a c. 400 km range extension from previous reports near Borroloola and Darwin. Discussion Three of the observed species, the Two-spotted Line-blue Nacadnba biocellata, Long- tailed Pea-blue and Fiery Jewel, are uncommonly reported from any location in monsoonal north-western Australia (Braby 2000), being more widespread in inland, southern and/or eastern Australia. The Two-spotted Line-blue has been previously reported in the region from only the vicinity of Darwin and in the catchment of the Prince Regent River in Western Australia. However, our four new localities suggest that it may be much more widespread, as suggested by Braby, with its small size resulting in it being readily overlooked. The Fiery Jewel has previously been reported in the Northern Territory from several locations in the vicinity of Darwin, and once each from the vicinity of Katherine and in east Arnhemland. The single, striking individual reported here from Kakadu National Park was observed from about 1 metre for several minutes as it perched on the foliage of an understorey sapling in tall woodland of Darwin Stringybark Eucalyptus tetrodonta and Darwin Woollybutt E. miniata that had not been burnt for several years. The Long-tailed Pea-blue is more widely dispersed in the region than the previous two species, but nevertheless uncommonly so. In addition to the report in Table 1 from Gove Peninsula, DF observed it closer to previously reported occurrences, at El Questro Station in Western Australia, where two individuals were seen and one netted in green grass near the bank of the Pentecost River. The remaining 19 species are widespread in parts of monsoonal north-western Australia (Braby- 2000), and most are fairly common to abundant in the Darwin area (pers. obs.). The records presented here are thus not particularly surprising. Some, however, may provide basis for more comprehensive generalisations about distributions. The numerous locations for, and frequent abundance of the Orange Ringlet Hjpocysta adian/e , Spotted Pea-blue Euchrysops cnejns and Small Dusky-blue Candalides crimes , including many records not listed here because they were in or close to previously reported occurrences (Braby 2000), suggest when combined with Braby's distribution maps that these species occur more or less throughout the tropical savannas of north-western Australia. At many of the sites where we observed the Small Dusky-blue, its food plants — several species of dodder-laurel (Cassy/hd) — were also prevalent. Numerous sightings of the Chocolate Argus suggest a similarly widespread distribution for the Top End of the Northern Territory at least. 4 Northern Territory Naturalist 18 (2005) D. Franklin et al. Table 1. New location records for some Top End and Kimberley butterflies. Checklist order, scientific nomenclature, and common names follow Braby (2000). NT = Northern Territory; WA = Western Australia; VRD = Victoria River District (NT). Bracketed dates are inclusive periods where dates of individual sightings were not noted. Notes enclosed thus " " are extracts from DF's diary. Family / species / location Date Notes Papilionidae Cressida cressida (Clearwing Swallowtail) NT: Gove Pen. - Daliwuy 12°21'S, 136°55'E Nymphalidae Melanitis leda (Evening Brown) (10-16)/7/'04 ''common in woodland" WA: El Questro Stn. - Amalia Gorge 14/8/2004 "sheltering on ... ground 15°57'S, 128°02'E Nymphalidae Mycalesis sirius (Cedar Bush-brown) ... in heavy shade" NT: Central Arnhem Rd 145 km from Nhulunbuy c. 12°45'S, 136°14'E 9/7/2004 one, in creekline grass NT: Liverpool R. crossing, Oenpelli- ManingridaRd 12"21'S, 134WE Nymphalidae Ypthima arctous (Dusky Knight) 20/7/2004 one, in floodplain grass WA: Ellenbrae Stn 15°59'S, 127°04'E 3/8/2004 one WA: Barnett River Gorge 16°32'S, 126°07'E Nymphalidae Hypocysta adiante (Orange Ringlet) 13/8/2004 one NT: Lorella Springs Stn Hstd 15°44'S, 135°39'E 12/6/2004 moist grass near creek NT: Limmen Nat. Pk - Butterfly Springs 14/6/2004 in moist grass near 15°38’S, 135°28'E creekline NT: Central Arnhem Rd, 145 km from Nhulunbuy c. 12°45'S, 136°14’E 9/7/2004 one, grass near creek NT: Central Arnhem Road at Flat Rock Ck c. 12°55'S, 135°20'E 17/7/2004 one, grass near creek WA: Barnett River Gorge I6°32’S, 126°07'E 13/8/2004 WA: Bamboo Ck, Gibb River Rd. 15°53'S, 127°21'E 13/8/2004 creekline Nymphalidae Polyura sempronius (Tailed Emperor) NT: Victoria River Roadhouse 15°37'S, 131°07'E 14/8/2003 Nymphalidae Cethosia penthesilea (Orange Lacewing) NT: Ramingining area c. 12°15'S, 134°58'E Nymphalidae Junonia hedonia (Chocolate Argus) c. 6/7/2004 NT: Wollogorang Stn - Settlement Ck 10/6/2004 2 or 3 in one creekside 17°13'S, 137°56'E location NT: Lorella Springs Stn Hstd 15°44’S, 135“39’E (12-13)/6/'04 NT: Roper Bar 14°43'S, 134°31'E 16/6/2004 "fairly common ... upper ... floodplain forest" NT: Gove Pen. - Gayngaru 12°11'S, 136°47'E 11 & 14/7/'04 "locally very common" WA: Adcock Gorge 16°55'S, 125°46'E Nymphalidae Tirumala hamata (Blue Tiger) 12/8/2004 "one at swampy creek" NT: Gove Pen. nr Garanhan 12°20'S, 136°56'E 10/7/2004 "one, vine-thicket" continued on next page New butterfly records Northern Territory Naturalist 18 (2005) 5 Nymphalidae Danaus affinis (Swamp Tiger) NT: Limmen Nat. Pk - Towns R. fishing camp 15°02'S, 135°14'E NT: Gove Pen. - Daliwuy 12°21'S, 136°55'E NT: Gove Pen. - Gayngaru 12°1TS, 136°47'E Lycaenidae Hypochrysops ignita (Fiery Jewel) NT: Kakadu Nat. Pk - 25 km NE of Munmarlary 12°21'S, 132°40’E Lycaenidae Arhopala centaurus (Purple Oak-blue) NT: Gove Pen. - Daliwuy 12°21’S, 136°55’E NT: Gove Pen. - Gayngaru 12°11 ’S, 136°47'E Lycaenidae Candalides erinus (Small Dusky-blue) NT: Elsey Nat. Pk - Mulurark 14°57'S, 133°13'E NT: Mataranka - Elsey Cemet. 15 ‘04'S, 133 '07'E NT: Elsey Nat. Pk - Salt Ck 15°01'S, 133°14‘E NT: Caranbirini Cons. Res. 16°17'S, 136°05'E NT: Roper Bar 14°43'S, 134°29'E NT: Gove Pen. - Daliwuy 12°21'S, 136°55'E NT: Gove Pen. - Gayngaru 12°11’S, 136°47'E WA: Ellenbrae Stn 15°59'S. 127°04'E WA: Mitchell Plateau c. 14°45'S, 125"40'E Lycaenidae Nacaduba biocellata (Two-spotted Line- NT: Jasper Gorge, VRD 16°02'S, 130°41'E WA: Hidden Valley 15°46'S, 128°45'E NT: nr Pine Creek 13°49'S, 131°50'E WA: El Questro Stn 15°57'S, 128°02'E Lycaenidae Prosolas dubiosa (Purple Line-blue) NT: nr Jasper Gorge, VRD 16“00'S, 130°39'E 15/6/2004 (10-16)/7/'04 (11 -14)/7/'04 28/8/2004 (10-16)/7/'04 (10-16)/7/'04 2/5/2004 2/5/2004 3/5/2004 6/6/2004 16/6/2004 (10-16)/7/'04 (11-14)/7/'04 3/8/2004 (5-6)/8/'04 blue) 3/5/2003 15/8/2003 8/6/2003 16/8/2004 4/5/2003 Lycaenidae Catopyrops florinda (Speckled Line-blue) WA: Ellenbrae Stn 15 C 59'S, 127°04’E 3/8/2004 Lycaenidae Lampides boeticus (Long-tailed Pea-blue) NT: Gove Pen. - Gayngaru 12°11'S, 136°47'E 14/7/2004 Lycaenidae Zizeeria karsandra (Spotted Grass-blue) NT: Jasper Gorge, VRD 16°02'S, 130°41'E 3/5/2003 Lycaenidae Zizina labradus (Common Grass-blue) NT: Nhulunbuy & Gayngaru 12°11'S, 136°47’E (11-14)/7/'04 Lycaenidae Famegana alsulus (Black-spotted Grass-blue) NT: Jasper Gorge, VRD 16°02'S, 130°41'E 3/5/2003 NT: nr Wollogorang Stn Hstd 17°13'S, 137°57'E 10/6/2004 NT: Gove Pen. - Daliwuy 12°21'S, 136°55'E (10-16)/7/'04 Lycaenidae Euchrysops cnejus (Spotted Pea-blue) NT: nr Wollogorang Stn Hstd 17°13'S, 137°57'E NT: Lorella Springs Stn 15°44'S, 135°39’E NT: Roper Bar 14°43’S, 134 0 3VE NT: Gove Pen. - Daliwuy 12°21'S, 136°55'E 10/6/2004 (12-13)/6/'04 16/6/2004 (10-16)/7/'04 Lycaenidae Freyeria putli (Jewelled Grass-blue) NT: Jasper Gorge, VRD 16°02'S, 130°41'E 3/5/2003 two together, near river bank a few a few one one one a few "several... sandstone" "abundant ... creekline'' abundant abundant a few in creekbed one one lawn & grassy area "fairly common" locally abundant "common" 6 Northern Territory Naturalist 18 (2005) D. Franklin etai. Records from northern Arnhemland of the Cedar Bush-brown Mycalesis sinus , Orange Ringlet and Orange Lacewing Cetbosia peuthesi/ea suggest continuity across the north coast of the Northern Territory between previously reported populations in the north¬ western Top End and at Nhulunbuy. Records of the Orange Ringlet in Western Australia may indicate continuity between previous reports from the central-west and east Kimberley, especially as DF also observed the species at other intermediate locations closer to the previously reported populations, at Adcock Gorge and El Questro Station (see Table 1 for coordinates) and Manning Gorge (16°38'S, 125°55'E). Records of the Purple Line-blue Prosotas dubiosa. Spotted Grass-blue Zigeeria karsandra , Black-spotted Grass-blue Famegana alsulus and Jewelled Grass-blue Freyeriapntli in the Victoria River District max' indicate continuity between previously- reported populations in the Kununurra area of Western Australia and those of the north-western Top End. In the course of three years of active field study of butterflies around Darwin, neither we nor any colleagues have obtained records of species new to the area. However, upon travelling to remoter areas, new location records were very readily obtained; metaphorically, we tripped over interesting records. This indicates a steep decline in the intensity’ of previous surveys away from the main settled areas, and that very much remains to be learned about butterfly distributions in the larger portion of north¬ western Australia. It also indicates that skilled observers, whether amateur or professional, can contribute greatly to our geographical understanding of butterflies simply by visiting and surveying remoter areas, carefully self-vetting records for accuracy, and then placing interesting observations on the public record. Acknowledgements Our joy in developing an acquaintance with the butterfly fauna of monsoonal Australia has been shared with Heather Ryan, Deb Bisa, Arthur & Sheryl Keates, Kim McLachlan and occasionally other people, and interactions with these observers have greatly enhanced our identification skills. Soren Faurbv, Arthur & Sheryl Keates and Niven McCrie assisted with the survey of Elsey National Park. Christine Maas drew DF's attention to the most exciting of the records presented here, the Fiery Jewel. The survey of Elsey National Park was conducted under Parks & Wildlife pennit no. 17457. References Barrett G., Silcocks A., Barry’ S., Cunningham R. and Poulter R. (2003) The New Atlas of Australian Birds. Royal Australasian Ornithologists Union, Hawthorn East, Vic. Blakers M., Davies S.J.J.P. and Reilly P.N. (1984) The Atlas of Australian Birds. Melbourne University Press, Carlton, Vic. Braby M.F. (2000) Butterflies of Australia. Their Identification, Biology and Distribution. CSIRO, Collingwood, Vic. New butterfly records Northern Territory Naturalist 18 (2005) 7 Braby M.F. (2004) i 'he Complete Field Guide to butterflies of Australia. CS1RO, Collingwood, Vic. Common I.F.B. and Waterhouse D.F. (1981) butterflies of Australia. Angus & Robertson, London. Dunn K.L. and Dunn L.E. (1991) R errin' of Australian butterflies: Distribution, lJfe History and Taxonomy. K.L. & L.E. Dunn, Melbourne. Franklin D.C., Woinarski J.C.Z. and Noske R.A. (2000) Geographic patterning of species richness among granivorous birds in Australia. Journal of btogcograpby 27, 829-842. Griffioen P.A. and Clarke M.F. (2002) Large-scale bird-movement patterns evident in eastern Australian atlas data, limn 102, 99-125, Grund R. and Hunt L. (2001) Some butterfly observations for the Kimberley andTanami regions. Western Australia. Victorian Entomologist 31, 19-23. Puccetti M. (1991) Butterflies of Doomadgee - northwestern Queensland. Victorian Entomologist 21, 142-147. Wilson C.G., Barrow P.H. and Michell C.R. (2003) New locations and host plants for Leichhardt's Grasshopper Pe/asida ephippigera White (Orthoptera: Pyrgomorphidae) in the Northern Territory. Australian Entomologist 30, 167-176. Sightings of the Long-tailed Pea- blue / Mtnpides boefiats (left, Max Mace) and Swamp Tiger Danans afftttis (below, Don Franklin) near the Gulf of Carpentaria are new locations for these species. Northern Territory Naturalist (2005) 18: 8-13 Do predators affect the survival of Macaria pallidata larvae? Implications for biological control of Mimosa pigra in the NT Blair Grace Weed Management Branch, Department of Natural Resources, Environment and the Arts, PO Box 30, Palmerston, NT 0830. Email: blair.grace@nt.gov.au Abstract There have been few studies investigating whether predators can affect the survival of insects that have been introduced into new regions. To address this, ants and birds were excluded from mimosa {Mimosa pigra) plants that had larvae of a leaf-feeding geometrid moth, Macaria pallidata placed on them. The moth is used as a biological control agent against mimosa in the Northern Territory. More larvae were observed when ants and birds were excluded. The ants present were generalists, probably attracted to mimosa by the nectar it supplies at the base of the leaves. Introduction Few studies have investigated factors influencing the establishment of insect populations introduced into new areas. Failure of an insect to colonise a new area is commonly attributed to climatic variables, often without an)' data to support such a claim (Clarke 2001). The actual causes of failure are rarely studied, but can have important implications for biological control programs (Day et al. 2004). Such knowledge should be used when selecting future biological control agents, considering the large costs associated with finding, testing and introducing them. The macaria moth Macaria pallidata (Lepidoptera, Geometridae) is a biological control agent released against the weed mimosa ( Mimosa pigra L.) in Australia. Macaria was identified as a potential biological control agent against mimosa, but was originally ignored because it was considered too vulnerable to predators and parasitoids (Flarley etal. 1995). Female macaria lay eggs on leaves and stems, and larvae feed for c. 13 days. Larvae are soft-bodied, slow moving “looper” caterpillars that grow up to 2 cm long and feed externally on mimosa leaves (Heard et al. 2001). Larvae go through five instars tiien form prepupae, which are obviously shorter, thicker and darker coloured. Most larvae descend to the ground to pupate. Larvae drop on a silken thread when disturbed (Heard et al. 2001, B. Routley, pers. comm.). Survival of Macaria larvae Northern Territory Naturalist 18 (2005) 9 Mimosa has been the target of a large, ongoing biological control program since 1979 (Paynter & Flanagan 2002). It is an invasive weed that infests approximately 800 km 2 of Top End floodplains, and has the potential to spread throughout much of tropical Australia (Walden et al, 2002). To date, twelve species of insects and two species of fungi have been released against this weed. Seven of these insect species have established (N. Ostermeyer, unpubl.). Macaria was the first insect released that primarily attacks mimosa leaves. Mimosa in the Northern Territory now grows more slowly and produces considerably less seed than it did before biological control (Paynter & Flanagan 2002, Paynter in press). Initial releases of macaria in Australia (between June 2002 and December 2004) appeared to be unsuccessful - no insects could be found at release sites, despite extensive searches (B. Roudey, unpubl.). Predators such as ants, spiders and birds have been observed eadng and disturbing caterpillars in the Top End, and birds have also been observed eating macaria adults. This experiment aimed to determine if ants and bird predators do influence survival of macaria larvae. Methods The experiment was conducted near Beatrice Hill Lagoon (12" 33’ S, 130° 18’ E), on the Adelaide River floodplain. The site was previously dominated with spike-rushes (Ekocharis spp.) (Story 1969), and much of the area was invaded by mimosa in the 1970s (Braithwaite et al. 1989). Since 1989 the inundated areas have been taken over by the introduced pasture grasses Hymenacbne amplexicanlis and hrachiaria mutica , leaving little Eleocbam remaining. Mimosa planted in 1999 was used for the experiment. These plants had been planted for previous experiments, and were spaced 3 m apart. To make plants more homogeneous and improve the probability of finding insects, plants were trimmed to c. 1.5 m high one month before the experiment commenced. To ensure the ant exclusion treatment was effective, all vegetation touching each plant was removed. Four treatments were applied randomly to mimosa plants: 1. Ants were excluded by manually removing all ants seen, then applying sticky gel (“Tac-gel”, Rentokil) around the base of each plant. This gel was inspected several times, and any large sticks or leaves removed. There were six replicate plants used in this treatment. 2. Birds were excluded with commercial bird netting, which was set up around the plants but not touching foliage. It was not known which bird species were likely to eat larvae (eight replicate plants). 3. Plants had both ants and birds excluded (seven replicate plants). 4. A control group was left untouched (seven replicate plants). 10 Northern Territory Naturalist 18 (2005) B. Grace Fifty larvae (third to fifth instar) were placed randomly on the foliage of each plant on 7 November 2002. Larvae on each plant were counted for 5 minutes on 8, 11 and 13 November. The experiment was terminated on 13 November, when no larvae were observed and all larvae would have pupated. Two observers did all the larvae counting, and were assigned plants at random. To test for differences between observers, both observers counted larvae on 12 plants on 8 November. A sample of ants from each plant was collected on 6 November and identified soon after. There was no specific survey of bird species present in the area. The number of larvae counted on 8 and 11 November were compared between treatments using generalised linear models with Poisson error distribution, after first removing effects of the person counting larvae by treating this as an additional variable, and checking for overdispersion. Results One day after placing larvae on plants, more larvae were found on plants where ants had been excluded (Figures 1, 2, d.f. = 1, % 2 — 32, p < 0.0001). Excluding birds also affected larval survival (d.f. = 1, % 2 — 3.8, p = 0.009). There was no important interaction between excluding birds and excluding ants (d.f. = 1, X 2 ~ 1.0, p = 0.3). 16 c (0 CD Q. T3 CD +-« c 3 o o 0) (B 12 - 8 4 ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ 0 Ants Ants + Birds Birds Predators excluded ♦ ♦ ♦ ♦ Control Figure 1. Number of larvae counted per plant, one day after larvae were placed on mimosa plants. Plants had either no predators excluded (control) or ants, birds, or both excluded. Raw data points (♦) are staggered on the x-axis so all data points can be shown, and horizontal bars show the mean for each treatment. Survival of Macaria larvae Northern Territory Naturalist 18 (2005) 11 50 l Figure 2. Number of larvae counted per plant, over time, after 50 larvae were placed on each plant at day 0. Plants had either ants (A), birds (•), both (■), or no predators excluded (X). Excluding ants increased the number of larvae found (d.f. = 1, y} = 20, p < 0.0001) four days after placing larvae in the field (Figure 2). Excluding birds appeared to slightly affect larv al survival (Figure 2), although this was not significant (d.f. = 1, % 2 — 2.9, p = 0.08). There was no important interaction between excluding birds and excluding ants (d.f. = 1, y} = 0.9, p = 0.3). Numbers of larvae observed decreased over time, with more larvae observed on plants with ants excluded (Figure 2). No larvae were observed on day 6. Overall, few larvae were found, and ants were observed disturbing larvae, which then descended to the ground on a silk thread. There were differences between observers (d.f. = 1, y} — 4.4, p = 0.03), which suggests that the larval counting process may not have been perfect and not all larv ae present were counted. Only one larva was found caught in the sticky gel and 21 prepupae were found; 12 in the ant-excluded plants, and 9 in plants that had ants and birds excluded. The ants collected were Polyrbachis cnw’leyi, P. scbenkii , P. sp. nr obtusa, RJjytidopouera sp. nr aurata , Odontomachus sp. nr lumen, Crematogaster sp. (C. laemceps group) and Ochetellus sp. Very little is known about the ecology of these species, but all ants appear to be generalist scavengers rather than specialised predators (A. Andersen, pers. comm.). Within two days of excluding ants, plants had obvious large, sticky globules exuding from the extrafloral nectaries at the base of each leaf. 12 Northern Territory Naturalist 18 (2005) B. Grace Crimson Finches Neochmia phaeton were observed near the mimosa plants, and twice were found trapped inside the bird nets; these are however primarily seed-feeders (Todd et al. 2003). The only insectivorous birds observed in the area were Rainbow Bee-eaters Merops ornat/is. Discussion Ants commonly disturbed macaria larvae, and either removed them or caused them to drop on silk threads. Birds had an effect on larval numbers, but this was not as important as the effect of ants. Low numbers of larvae were found after release. It is possible that many larvae were present in the foliage but not found. The cryptic colour of the larvae and differences in larvae counted on the same plants Ire tween observers suggests this is likely. Many larvae probably also pupated within several days of being placed on plants. Ant species found on mimosa were generalists. Given that large sticky globules were observed at the base of the leaves when ants were excluded, we can conclude that ants feed from these extrafloral nectaries. It is likely that mimosa evolved these nectaries to attract such ants, which then deter herbivores, as is common in the Mimosaceae outside of Australian rangelands (Norris et al. 1994); these associations appear to be common in the native range of mimosa. Of the leguminous plants studied in Mexico, 73% had close associations with ants, and use of extrafloral nectaries was the most common ant-plant interaction in the native range of mimosa (Rico-Gray et al 1998). This association with a range of ants that deter herbivores may be one reason why mimosa has become such a 'successful' weed. The sticky gel may have also deterred other predators such as spiders, frogs and lizards. Being a wetland, the area has an abundance of frogs, such as I Jtoria ///color, which commonly sits on mimosa branches (pers. obs.). These other predators were not investigated in this study', but none were observed caught in the gel. Overall, only a small proportion of all larvae placed on mimosa plants were seen. Larvae used in this experiment were late instar, so many may hav e dropped to the ground to pupate during the trial. Larvae are also cry ptic, and some may have been missed while counting. When this experiment was conducted, macaria had not been recovered from any site where it had been released. Recent surveys, however, have found that it has established and spread widely on mimosa in the Northen Territory (B. Routley, unpubl.). Although predators such as ants and birds do remove and disturb a considerable proportion of larvae, macaria is still capable of surviving and spreading. This may be due in part to the insect’s high fecundity (Heard et al 2001), or because the density' of mimosa thickets would allow larvae to simply drop onto a lower branch if disturbed. The effect predators have on population density is still unknown. This Survival of Macaria larvae Northern Territory Naturalist 18 (2005) 13 paper shows that predators can remove considerable numbers of larvae, although this does not necessarily prevent insects from colonising new areas. Acknowledgements Merrilyn Paskins, Andrea Wilson, Bruce Hitchins, Bert Lukitsch and Nadine Graham all helped set up and run the experiment. Alan Andersen kindly identified ants collected, and commented on earlier versions of this paper, as did John Woinarski and Bron Routley. This work was funded by the Department of Infrastructure, Planning and Environment and the Natural Heritage Trust. References Braithwaite R.W., Lonsdale W.M. and Bstbergs J.A. (1989) Alien vegetation and native biota in tropical Australia: The impact of Mimosapigra. biological Conservation 48, 189-210. Clarke, C.R. (2001) Climate matching in the colonisation of biological control agents against Chryscmthemoides momhfira and Marrubium ndgare. PhD Thesis. University of Adelaide. Day M.D., Briese D.T., Grace B.S., Holtkamp R.H., Ireson JSheppard AAV. and Spaflord J.H. (2004) Improving release strategies to increase the establishment rate of weed bioconlrol agents. Proc. 14th Australian Weeds Conference, Wagga Wagga. pp. 369-373. Harley K., GilletJ., Winder J., Porno I., Segura R., Miranda H. and Kassulke R. (1995) Natural enemies of Mimosa pigra and A 1. beriandkti (Mimosaceae) and progress for biological control of M. pigra. EnvironmentalEntomology 24, 1664-1678. Heard T.A., Mira A. and Zonneveld R. (2001) Application to release the defoliating Lepidoptera Macaria pallidata into Australia for biological control of the weed Mimosa pigra. Report submitted to Australian Quarantine and Inspection Sendee. Norris K.R., Farrow R.A. and Drake R.W. (1994) General Biolog)-. In Systematic and Applied Entomology: In Introduction, (ed. I.D. Naumann), pp. 68-108. Melbourne University Press, Melbourne. Pavnter Q. (in press). Evaluating the impact of a biological control agent Carmenta mimosa on the woody wedand weed Mimosa pigra in Australia. Journal of Applied Ecology. Pavnter Q. and Flanagan G.J. (2002) Integrated management of Mimosa pigra. Proceedings of the 13th Australian Weeds Conference, Perth, pp. 165-168. Rico-Gray V., Garcia-Franco J.G., Palacios-Rios M., Diaz-Castelazo C., Parra-Table V. and Navarro J.A. (1998) Geographical and seasonal variation in the richness of ant-plant interactions in Mexico. Biotropica 30, 190-200. Story R. (1969) Vegetation of the Adelaide-AUigator Area. In Inuds of the Adelaide-Alligator Area, Northern Territory. Land Research Series No. 25, CSIRO, Melbourne. Todd M.K., Felton A. and Garnett S.T. (2003) Morphological and dietary differences between common and uncommon subspecies of Crimson Finch, Neochmia phaeton , and Star Finch, Neochmia mjicauda, in northern Australia. Emu 103, 141-148. Walden D., van Dam R., Finlayson M., Storrs M., Lowry J. and Kriticos D. (2002) A risk assessment of the tropical wetland weed Mimosa pigra in northern Australia. Proceedings 3rd International Symposium on the Management of Mimosa pigra. Darwin, pp.11-21. Northern Territory Naturalist (2005) 18: 14-20 Threatened fishes of the Northern Territory Helen K. Larson Museum and Art Gallery of the Northern Territory, PO Box 4646, Darwin, NT 0801. Email: helen.larson@nt.gov.au Abstract Fifty species of Northern Territory fishes have threatened species status according to several Acts and organisations. Almost half of these species (26) are elasmobranchs, of which 11 are whaler sharks (family Carcharhinidae). Nine other listed species are syngnathids (seahorses and pipefishes), which are speciose in the Northern Territory'. Most of the different threatened species listings assign different levels of threat, although the two speartooth sharks, the Finke desert-goby, the four sawfish and the Angalarri grunter all share quite similar listings between the ASFB, EPBC and PWCNT. The poor state of knowledge of die Territory’s fish populations has hindered the assessment processes. Introduction The Northern Territory has a diverse but not well-known fish fauna, occurring in desert waterholes to mangrove creeks and offshore over the continental shelf. About 1300 fish species are known from the Northern Territory' (NT), a number of which are undescribed (H. Larson, in prep.). New fish species are still turning up in Darwin Harbour (e.g. Larson and Takita 2004), and a long-term study of coral reefs of the Cobourg Peninsula Marine Park is revealing many more fish records (V. Gomelyuk videotapes). There are few recent publications which summarise portions of the NT fish fauna, examples being Taylor's 1964 Fishes of Arnhem I miu! (the results of a 1948 American - Australian scientific expedition to eastern Arnhem Land), Larson and Martin’s 1990 Freshwater Fishes of the Northern Territory and Larson and Williams’ 1997 checklist of Darwin Harbour fishes. The fish fauna is still not fully documented, and the status of the fish populations themselves is even less well-known, with the exception of some commercially important species such as Barramundi 1 Mtes calcarfer and Spanish Mackerel Scomheroides commersonianns. The first conservation listing of NT fishes was in Harris (1985) (the proceedings of the Australian Society for Fish Biology’s (ASFB) first conference on Australian threatened fishes). This included only two NT species: the Exquisite Rainbowfish Melanotaenia exquisita and Midgley’s Grunter Pingalla midgleyi, which were both listed as having restricted distributions. Provided below are the various national and international species listings and some comments on the present status of NT fish species. NT threatened fishes Northern Territory Naturalist 18 (2005) 15 Remarks Since Harris’ listing, additional research has added to our knowledge of NT fish distributions and the number of threatened species. Northern Territory threatened fish are now listed by the Australian Society' for Fish Biology’s Threatened Fish Committee (ASFB 2003), the International Union for the Conservation of Nature’s Red List of Threatened Species 2003 (IUCN; http://www.redlist.org/), Pogonoski et al. (2002), the Commonwealth Government (under the Environment Protection and Biodiversity Conservation Act 1999) and the Parks and Wildlife Commission of the NT (PWCNT 2002; also online). All NT species with a threatened status under any of these bodies are shown in Table 1 (these are currently under review). The IUCN and ASFB list the greatest number of threatened NT species. Out of the total of 50 listed species, IUCN lists 49 and ASFB lists 38 species. They both agree that 14 of these (mostly marine) species are Data Deficient. The EPBC Act protects fauna by listing species and communities within Commonwealth areas as well as protected (conservation) areas and is triggered by certain actions (e.g. a development proposal for an area inhabited by listed species). Listed fishes are afforded protection within Commonwealth waters, while the NT has its own legislation designed to protect fauna. Pogonoski et al.’ s (2002) “Conservation Overview and Action Plan for Australian Threatened and Potentially Threatened Marine and Estuarine Fishes” was simultaneously published in softcover and on the Environment Australia website in February' 2002 (http://www.deh.gov.au/coasts/species/marine-fish/index.html). Pogonoski et al. (2002) includes an overview in some detail of the conservation status of 114 species of marine and estuarine fishes and incorporates information provided by the ASFB threatened fish committee as well as the results of a specialist workshop on Australian threatened fish populations. Under the Territory Parks and Wildlife Act 2000 , the PWCNT produced a list of threatened fauna of the NT, which included nine threatened fish species (three endangered and six vulnerable), in addition to listing 188 species as Near Threatened, Least Concern or Data Deficient (PWCNT 2002). The PWCNT list of threatened species is not alvvay'S congruent with IUCN, EPBC or ASFB listings (Table 1). For example, the Freshwater Tongue Sole Cynoglossus heferolepis is an estuarine to freshwater species known from the Adelaide River to the East Alligator River and southern New' Guinea (Allen 2001), that is not endangered or threatened, but was apparently listed by the PWCNT based on information in Allen et al. (2002). The main consensus between the ASFB, EPBC and PWCNT listings seems to be the two species of speartooth sharks Glypbis sp. A and sp. C, Finke Desert-goby Cblamydogobius japalpa. Grey Nurse Shark Carcharias taunts, the four sawfish Pristis and Anoxypristis, and the Angalarri Grunter Scortnm neili. The PWCNT’s Strategy for the Conservation oj Threatened Species and Ecological Communities in the Northern Tenitmy oj Australia is a management tool for dealing with threatened taxa, but it docs not mention aquatic species or fish. 16 Northern Territory Naturalist 18 (2005) _ H. Larson Table 1. Fish species of conservation significance in the NT. 1UCN = 2002 IUCN Red List of Threatenened Species; EPBC = Environmental Protection and Biodiversity Conservation Act 1999; COAP = Pogonoski et al. (2002); ASFB = Australian Society for Fish Biology (2003); PWCNT = Parks and Wildlife Commission of the NT. CE = Critically Endangered, EN = Endangered, VU = Vulnerable, NT = Lower Risk (Near Threatened), LC — Lower Risk (Least Concern), LR = Lower Risk (Conservation Dependent), DD = Data Deficient. Species listed as Data Deficient by PWCNT (2002), but not elsewhere, are not included. Species IUCN EPBC COAP ASFB PWCNT Ariidae Cinetodus froggatti Small-mouthed Catfish DD DD Atherinidae Craterocephalus centralis Finke River Hardyhead NT NT Carcharhinidae Carcharhinus amblyrhynchos Grey Reef Shark NT LC C. brevipinna Spinner Shark NT LC - - C. falciformis Silky Shark LC LC - - C. leucas Bull Shark NT LC - - C. limbatus Blacktip Shark NT DD DD - C. obscurus Dusky Shark NT NT NT - C. plumbeus Sandbar Shark NT NT NT - Galeocerdo cuvier Tiger Shark NT LC - - Glyphis sp. A Speartooth Shark CE CE CR CE EN G. sp. C Northern River Shark CE EN EN EN EN Triaenodon obesus Reef White-tip NT LC - - Cynoglossidae Cynoglossus heterolepis Freshwater Tongue Sole EN Dasyatidae Himantura chaophraya Freshwater Whipray VU VU VU DD Taeniura lymma Blue-spotted Fantail Ray NT LC NT - Urogymnus asperrimus Porcupine Ray VU NT NT - Eleotridae Hypseleotris sp. Katherine River Gudgeon DD . DD Engraulidae Thryssa scratchleyi Freshwater Anchovy DD DD Gobiidae Chlamydogobius japalpa Finke Desert-goby VU VU VU Silhouettea hoesei Hoese's Silhouette Goby DD - DD - Mobulidae Manta birostris Manta Ray DD LC - - continued on next page NT threatened fishes Northern Territory Naturalist 18 (2005) 17 Table 1 continued Species IUCN EPBC i COAP ASFB PWCNT Myliobatidae Aetobatus nan'nan White-Spotted Eagle Ray DD - LC - - Odontaspidae Carcharias taurus Grey Nurse Shark VU CE EN EN DD Orectolobidae Orectolobus ornatus Banded Wobbegong LC (E coast) DD DD Pristidae Pristis microdon Freshwater Sawfish CR VU - CE DD P. clavata Dwarf Sawfish EN - - EN VU P. zijsron Green Sawfish EN - - EN VU Anoxypristis cuspidata Narrow Sawfish VU - - VU VU Rhincodontidae Rhincodon typus Whale Shark VU VU DD DD _ Rhinidae Rhynchobatus australiae White-Spotted Shovel-Nose Ray VU LC Serranidae Cromileptes altivelis Barramundi Cod LR LR Epinephelus lanceolatus Queensland Grouper LR - - LR - E. coioides Estuary Rockcod LC - - LC - E. fuscoguttatus Flowery Rockcod LC - - LC - E. malabaricus Malabar Grouper LC - - LC - Sphyrnidae Sphryna lewini Scalloped Flammerhead NT - LC " - S. mokarran Great Flammerhead DD - LC - - Syngnathidae Doryrhamphus dactyliophorus Banded Pipefish LC - - LC - Hippichthys parvicarinatus Estuary Pipefish LC - - LC - Hippocampus alatus Winged Seahorse DD - DD DD - H. dahli Low-Crowned Seahorse NT - - NT - H. multispinus Northern Spiny Seahorse DD - DD DD - H. taeniopterus Yellow Seahorse DD - DD DD - Solegnathus hardwickii Pallid Pipehorse DD - - DD S. lettiensis Gunther’s Pipehorse DD - - DD - Syngnathoides biaculeatus Alligator Pipefish DD - - DD Terapontidae Pingalla lorentzi Lorentz' Grunter VU P. midgleyi Midgley’s Grunter NT - - NT - Scortum neili Angalarri Grunter NT - - NT VU Xiphiidae Xiphias gladius Broadbill Swordfish DD - DD DD 18 Northern Territory Naturalist 18 (2005) H. Larson The seahorses, genus Hippocampus (Table 1), are all IUCN Red-listed (one species listed as Endangered, 20 Vulnerable and 11 Data Deficient). In addition, there is considerable confusion over the correct names for each species. No revision of the whole genus has yet been carried out, although Kuiter (2001) attempted to identify and name the Australian species. There are 33 Hippocampus species presently listed by the IUCN. Five seahorse species occur in the NT, but none of the names currently used for the NT species appear on the IUCN list, due to the nomenclatural confusion. The species are concealed and confused with other names on the list, and the three recently described NT species (Kuiter 2001) do not appear on the IUCN list. Work is continuing on Hippocampus species by Sara Lourie of McGill University, and it is hoped that this charismatic group of fishes will soon be better understood. Of the 51 species listed in Table 1, 26 are elasmobranchs (sharks, rays and sawfish). These large and often slow-growing fishes have been recognised as being highly vulnerable to fishing as a threatening process, either as bycatch or as target species (Pogonoski et al. 2002). Some of these are large and charismatic; for example the Freshwater Sawfish Pristis microdou , which reaches about 4.6 m (Allen et al. 2002), and is Australia’s largest freshwater fish, famous for waiting out the dry season in remote areas in small waterholes. Of the listed elasmobranchs, 11 belong to the family Carcharhinidae (whaler sharks). This family includes not only the speartooth sharks ( Glyphis sp. A and C) and the Tiger Shark Galeocerdo tuvieri, but also the various blacktip and grey whalers which are taken commercially in NT waters. In recent years there has been considerable effort expended to determine the status of various elasmobranch populations, resulting in a number of species receiving IUCN Red List status. The undescribed Glyphis species are both listed as Critically Endangered by the IUCN (2002), and as Endangered under the Territory Parks and Wildlife Conservation Act 2000, but no NT management program is in preparation (PWCNT 2002). A national Recover)' Plan for Glyphis is currently being drafted. Glyphis sp. A and Glyphis sp. C are known to inhabit several rivers in the NT, but their exact distributions and population numbers are not known. Museum and Art Gallery of the Northern Territory records and sight records of visiting researcher Tim Berra (in litt.) indicate that juvenile Glyphis sp. C may be common in the Adelaide River. The 25 non-elasmobranch Near-Threatened category' fish in Table 1 are mostly commercially significant (e.g. groupers, rock cods), syngnathids (seahorses and pipefishes) or small freshwater species (e.g. grunters). Some freshwater fish have very restricted distributions, making them vulnerable to changes in water quality, flow, condition or competition from introduced species. For example, the Finke River Hardyhead Craterocephalus centralis and the Finke Desert-goby Chlamydopobius japalpa are known only from the Finke River system in central Australia, and the hardyhead is the only species of the genus in that system. Both these species are highly' vulnerable to the potential effects of any introduced species (which have been found in waterways not far from the Finke system), such as Gambnsia holbrooki (the inappropriately-named mosquitofish) recendy removed from Ilparpa Swamp and the Murray Cod NT threatened fishes Northern Territory Naturalist 18 (2005) 19 Maccullocbella peeli recently found to be illegally introduced into Policeman Waterhole in the Davenport Ranges. The NT has been fortunate so far in that most occurrences of introduced exotic or noxious fish species have been dealt with successfully or otherwise contained (e.g. the removal of Jewel Cichlids Hemichromis bimaculatus , from Ludmilla Creek at Fannie Bay) and that Northern Territory Fisheries has an active aquatic pest team to deal with such outbreaks. Terrestrial parks and reserves within the NT provide some protection to threatened fish species, as do the few marine protected areas. There are only three of the latter: two small Aquatic Life Reserves in Darwin Harbour (at Doctors’ Gully and East Point) and one large Marine Park (Garig Gunak Barlu National Park, previously known as Cobourg Marine Park). The Commonwealth-administered Kakadu National Park includes several major river estuaries and islands as well as a range of freshwater habitats. Taking of fish (by hook and line) is permitted in the National Parks, but not in the Aquatic Life Reserves. The Northern Territory Fisheries Act 1988 is intended for the management and conservation of marine fishes, utilising Fisheries Management Plans to “conserve, enhance, protect, utilise, and manage the fisheries of the Territory”. So far, fish species in the NT are only protected by bag limits on several species, commercial fishing closures and the difficulty of access to many habitats. No species are wholly protected under NT legislation other than groupers of the genus Epinepheh/s larger than 1.2 m long (this protects adult females of the Giant Grouper Epinephelns lanceolatus). References Allen G.R. (2001) Field Guide to the Freshwater Fishes of New Guinea. Publication No. 9, Christensen Research Institute, Madang. Allen G.R., Midgley S.H. and Allen M. (2002) Field Guide to Freshwater Fishes of Australia. Western Australian Museum, Perth. Australian Society' for Fish Biology. (2003) Conservation status of Australian fishes - 2003. Australian Society for Fish Biology Newsletter 33(2), 60-65. Harris J.H. (1985) (ed.) Proceedings of the Conference on Australian Threatened Fishes, Melbourne, 15-16 August 1985. Australian Society' for Fish Biology, Sydney. Kuiter R. (2001) Revision of the Australian seahorses of the genus Hippocampus (Syngnathiformes: Syngnathidae) with description of nine new species. Records of the Australian Museum 53,293-340. Larson H.K. and Martin K.C. (1990) The Freshwater Fishes of the Northern Territory. Handbook Series No. 1, Northern Territory Museum of Arts and Sciences, 1-102. Larson H.K. and Williams, R.S. (1997) Darwin Harbour fishes: a survey and annotated checklist. In Proceeding of the Sixth International Marine Biological Workshop. The Marine Flora and Fauna of Darwin Harbour, Northern Territory, Australia (eds. J.R. Hanley', G. Caswell, D. Megirian and H.K. Larson.), pp. 339-380. Museums and Art Galleries of the Northern Territory and Australian Marine Sciences Association, Perth. 20 Northern Territory Naturalist 18 (2005) H. Larson Larson H.K. and Takita T. (2004) Two new species of Periophthalmus (Teleostei: Gobiidae: Oxudercinae) from northern Australia, and a re-diagnosis of Periophthalmus mvaegttineaensis. The Beagle, Records of the Museums and Art Galleries of the Northern Tern'tor/ 20, 175-185. Parks and Wildlife Commission of the Northern Territory. (2002) Threatened species of the Northern Territory information package. Northern Territory Government, Darwin. Pogonoski J.J., Pollard D.A. and Paxton |.R. (2002) Conservation omriew and action plan for Australian threatened and potentially threatened marine and estuarine fishes. Environment Australia, Canberra. Taylor W.R. (1964) Fishes of Arnhem Land. In Records of the American-Anstralian Scientific Expedition to Arnhem Iaind. 4. Zoology (ed. R.L. Specht), pp. 45-307. Melbourne University Press, Parkville. The Northern River Shark Gfypbis sp. C occurs in several Top End rivers and is thought to be Endangered, perhaps critically so. (Helen Larson) Northern Territory Naturalist (2005) 18: 21-33 A survey of nocturnal reptiles of Robin Falls, Northern Territory: implications of Bufo marinus Lorrae J. McArthur and Jeanne E. Young School of Education, Health and Science, Charles Darwin University, Darwin, NT 0909 Abstract Twenty-two species of nocturnal reptiles were identified in the Robin Falls area during one year (2001-02) of regular observational surveys. More reptiles were observed during the build-up and early dry season compared to the wet and the dry seasons. The greatest number of individuals (42) and species (15) of nocturnal reptiles were recorded during the build-up (Sept.-Nov.), during which average night-time temperatures of 28"C, absolute humidity of 19.07g/m 3 , and total rainfall of 190 mm were experienced. While comparisons of seasonal climatic factors produced trends, temperature, humidity and rainfall did not show any significant relationship with the number of observed reptiles. Nearly one third of the nocturnal reptile species listed in the Robin Falls area are considered to be susceptible to the Cane Toad Bufo marinus. This study provides preliminary quantitative baseline data from which to examine any future detrimental effects of the Cane Toad on nocturnal species of reptiles in the Robin Falls area. Introduction Robin Falls is a small tourist haven, located near the Adelaide River Township, which lies about 120 km south of Darwin in the Northern Territory (NT). The falls are reached via Dorat Road, which turns off the Stuart Highway near the Adelaide River Township. A 15-minute walk from the Robin Falls car park leads along a creek edged with monsoon forest. The creek originates from a spring above the escarpment and carves its way through swamp and open woodland until it clears the edge of the plateau and plunges into a large rock basin in the valley below. After the valley, the creek passes underneath Dorat Road and heads towards lower floodplains. Dorat Road travels through plateau and lowland wetland and over many tributaries similar to that of Robin Falls, all of which feed into the Adelaide River. Plant communities present in this region include woodland, monsoon forest, swampland and floodplain. Robin Falls is privately owned, but graciously opened to visitors, and managed by the Landcare Council NT and other community groups such as Coomalie Water Watch and Greening Australia. There is little evidence of disturbance in the area except for a few small cleared areas for camping sites, a short gravel access road and a narrow 22 Northern Territory Naturalist 18 (2005) L. McArthur & J. Young 500 m walking track leading to the falls, and little invasion by weedy plant species. In recent years, fires have occurred evert' 1—2 years (fire events noted in 2000 and 2001). Dorat Road passes many privately owned properties that show litde evidence of development. There is a banana plantation not far from the Adelaide River Township but most properties appear unaltered, with little (generally pastoral) or no land use. A compilation of nocturnal species of reptile that have been identified (during 1911— 2002) in the Robin Falls area is available from the Biological Records Scheme (BRS), set up by the Parks and Wildlife Commission of the Northern Territory’ (PWCNT). However, comprehensive scientific research, such as long-term ecological studies, is lacking in this area. This is unfortunate given that the Cane Toad Bnfo marinas is slowly progressing towards Darwin. Cane Toads were introduced into Queensland in 1935 to act as a biological control for the beetle Dermolepia albohirtum , which is a pest in sugar cane crops (Straughan 1966). While ineffective at controlling the cane beetle, the Cane Toad has effectively invaded much of the north-eastern coast and hinterland of Australia. The toads possess large, paired parotid glands located dorsally on the neck, which produce a cocktail of toxic chemicals including biogenic amines and hallucinogenic compounds such as the bufogenins (Sutherland 1983, Edstrom 1992). The Australian native fauna have not co-evolved with these types of toxins, and anecdotal observations of species declines within areas of toad invasion and recent studies demonstrate that many Australian predators are adversely susceptible to these compounds (Covacevich & Archer 1975, Catling et at. 1999, van Dam et a/. 2002). Currently, efforts are being made to conserve local species, such as the Quoll Recovery Plan under the Island Arks Project, which aims to provide safe havens, free from the impacts of exotic species, for native species (B. Rankmore pers. comm.). At the time of this research there had been no records of B. marinus in the Robin Falls area (Frogwatch 2004). Dorat Road leading to Robin Falls is well known among Australian herpetologists for its reptile diversity. During one year of regular surveys in the Robin Falls area, we recorded observations of ail nocturnal species of reptile. In this paper, we summarise the number of reptiles and species observed by season, and present an analysis of the relationship between the weather conditions and the number of nocturnal reptile species observed in each season. A preliminary’ list of nocturnal reptiles in the Robin Falls area is also presented, indicating which local species may be impacted by the Cane Toad. Methods Visual encounter surveys (VES) were conducted along line transects between 21:00 and 24:00 hours every’ 10 days from December 2001 to November 2002 (Crump & Scott 1994). Surveys were conducted on two transects: Dorat Road and Robin Falls (13° 21’ 20”S, 131° 07’ 17”E). Robin Falls nocturnal reptiles Northern Territory Naturalist 18 (2005) 23 Dorat Road: the first transect was conducted along 15 km of bitumen road, starting from the Adelaide River township turnoff and ending at the Robin Falls turnoff. Reptiles were located from a car using headlamps and car headlights. The car traveled on average 40 km/hr, while the road and road shoulders were searched for reptiles. Robin Falls: the second transect used the 500 m long walking track that begins at the Robin Falls car park and ends at the falls. Reptiles were searched for on foot using headlamps. Walking pace was slow, and 2 m on either side of the walking track were searched. The VES were conducted at a consistent pace, and the focal area surveyed was within the predetermined radius to the left, front and right of the observer. The surveys did not involve shifting leaf litter or lifting and turning rocks and logs, thus allowing us to minimise habitat disturbance, to eliminate the risk of injury to the animal and observer from improper handling of animals, and to comply with the legal guidelines of the Parks and Wildlife Commission of the Northern Territory' (PWCNT) (it is a criminal offence to handle native wildlife unless holding a PWCNT permit). All nocturnal species of reptile encountered were recorded (amphibians were not included in this study because they are a focus of a separate paper). Air temperature (°C) and relative humidity' (%) were measured using a hand held meter (HM34C Vaisala), and rainfall was measured by a rain gauge. Analysis Four recognized seasons were used to compare species numbers; build-up (Sept.- Nov.); wet (Dec.-Feb.); early dry (Mar.-May); and dry (Jun.-[ul.). Results were collated to show the number of each species of nocturnal reptile found in each season, and statistically compared using ANOVA (Systat, SAS). Climatic variables and the numbers of species identified are presented graphically and correlated statistically using logarithmic regression, for poisson distributed count data (Systat, SAS). Since humidity is dependent on air temperature (Ta), for the purpose of statistical analyses relative humidity' was converted to absolute humidity (AH) (g/m 3 ) using the Smithsonian water density table at the given ambient temperature (List 1971). The level of significance for all analyses was set at p < 0.05. The list of reptile species that may be found in the Robin Falls area was compiled using information from Wilson and Swan (2003) on the geographical distribution of reptiles and from species listed in the BRS (1911-2002). Results Nocturnal species of reptiles observed in the Robin Falls area Table 1 provides a summary of the names and numbers of each species encountered on line transects in the Robin Falls area each season. A total of 22 nocturnal reptile 24 Northern Territory Naturalist 18 (2005) L. McArthur & J. Young species were recorded during one year of surveys (36 surveys). The Freshwater Crocodile Crocodjlits johmtoni, was encountered along Anniversary Creek (a tributary crossed by Dorat Road); however, the species was excluded from any analysis because the tributary was not consistently surveyed for this species throughout the survey period. Table 1. A list of the species and total numbers of each of the reptiles observed on Dorat Road and Robin Falls transects, summarised by season. Wet Early dry Dry Build-up Species (Dec.-Feb.) (Mar.-May) (Jun.-Aug.) (Sept.-Nov.) Gekkonidae Gehyra australis 3 1 2 4 Heteronotia binoei 2 Oedura marmorata 1 Strophurus ciliaris ciliaris 1 4 1 Pygopodidae Delma tincta 1 Lialis burtonis 3 5 Pythonidae Antaresia childreni 4 1 3 Liasis mackloti 1 Liasis olivaceus 1 3 Morelia spilota 1 Acrochordidae Acrochordus arafurae 15 Colubridae Boiga irregularis 1 2 Stegonotus cucullatus 1 Tropidonophis mairii 1 2 Elapidae Acanthophis praeiongus 13 Brachyurhophis roperi 1 Furina ornata 1 1 2 1 Pseudechis australis 1 2 Pseudonaja nuchaiis 1 Rhinoplocephalus pallidiceps 1 Suta punctata 1 Vermicella intermedia 1 Number of species 6 11 5 15 Total number of reptiles 8 30 10 42 Robin Falls nocturnal reptiles Northern Territory Naturalist 18 (2005) 25 Comparisons between season and number of observed nocturnal reptiles The numbers of nocturnal reptiles observed in each season were significandy different (ANOVA: F 332 = 9.538; p < 0.001). The greatest number of reptiles (42) and number of species (15) were found during the build-up (Sept.-Nov.) (Table 1). The number of species identified during the build-up included over half of the total species (22) observed during the one year of surveys. The number of individuals encountered during the early dry was also high (30), but was dominated by a large number (15) of Arafura File Snakes Acrocbordus arafurae found in Anniversary Creek. Despite the predominance of file snakes, the number of species observed during the early dry (11) included half the total number of species encountered during the entire survey period. Fisher’s PLSD post-hoc tests (5% significance level) found no statistical difference between reptiles observed during rhe build-up and the early dry (p = 0.1190). The lowest numbers of reptiles were observed during the wet (8) and the dry (10) seasons, which were not significantly different (Fisher’s post-hoc: p = 0.7912) between seasons. Despite the low numbers of animals observed during the wet, two species, the Northern Shovel-nosed Snake Bracbynrhophis rnperi and the Black-necked Legless Lizard Delma tincta, were encountered during this season only. Results from analyses of variance found significant seasonal differences in all climatic factors (ANOVA Ta: F 3>32 = 20.834, p < 0.001; AH: F 3 , 32 = 71.491, p < 0.001; Rain: F 3 ,32 = 5.486, p < 0.004). Night-time air temperature and absolute humidity were statistically higher in the wet, early dry and build-up than in the dry (Fisher’s post-hoc: p < 0.0001) (Figure 1). While the wet season had similar average temperature and humidity to that of the build-up and early dry, total rainfall (816 mm) was greater than all other seasons (Fisher’s post-hoc p < 0.0033) (Figure 1). Statistical correlations between climatic variables and the number of observed reptiles All climatic variables were log transformed to minimise variation either side of the mean and fit assumptions of the analyses. A correlation matrix (StatView, SAS) produced small positive relationships between reptile counts and log temperature and log absolute humidity (maximum R- value 0.315). Further analysis, using logarithmic regression of poisson distributed reptile count data, found no statistically significant (p > 0.05) relationship between the number of reptiles and climatic variables (Table 2). The statistical correlation between reptile number and log rainfall was not significant (R 2 = 0.027) and therefore was not included in the regression model. While no significant statistical correlations were found between climatic factors and the numbers of reptiles found in each season, trends are evident when the number of species observed in each season is considered along with the plot of the climatic variables for each season (Figure 1). The highest numbers of reptiles were found during warm, humid conditions, and the lowest numbers w T ere found in the coolest, driest season. 26 Northern Territory Naturalist 18 (2005) L. McArthur & J. Young Figure 1. The number of nocturnal reptile species identified in each season and associated climatic variables (ED = early dr)'; BU = build-up). Table 2. Reptile counts and climatic variables correlated using logarithmic regression (StatView, SAS). Predictor variables were log transformed, n = 36. Coefficient Standard error Standard coefficient R 2 F-Value p-Value log temperature 12.790 7.065 0.297 0.088 3.277 0.0791 log humidity 4.394 2.288 0.313 0.098 3.689 0.0632 A list of reptile species in the Robin Falls area A list of nocturnal species of reptiles reported to inhabit the Robin Falls area is presented in Table 3 (Wilson & Swan 2003, BRS, PWCNT). Half of the nocturnal reptile species (22 of 44) were recorded during one year of regular surveys (asterisks in Table 3). At least 50% of species within most families were recorded on Dorat Road and/or at Robin Falls. Most species of the Elapidae and Pythonidae were recorded (except for the Black-headed Python) but no species of Blind Snake (Typhlopidae) were identified, which is not unusual given their burrowing behaviour (Cogger 2000). Robin Falls nocturnal reptiles Northern Territory Naturalist 18 (2005) 27 Cane Toad toxins are considered life-threatening to a number of the listed repdles (Table 3), based on information provided on diet (Wilson & Swan 2003), studies on the effects of toxins on some species of snake (Covacevich & Archer 1975, Phillips et al. 2003) and an impact assessment report (van Dam et at. 2002). Nearly one third (30%) of nocturnal species of reptile arc considered susceptible to the Cane Toad. Only two out of the total species listed are known to be tolerant of Bufo toxins: the Keelback Snake Tropidonophis tnairii and the Slaty-grey Snake Stegonotm cucullatus (Ingram & Covacevich 1990, Phillips et al. 2003). All other reptiles listed do not appear to prey on amphibians. Discussion During this study, 22 species of nocturnal reptiles were identified near Robin Falls. Most species (15) were encountered during the build-up when humidity and temperature were highest. The number of species observed during the early dry was similarly high (11), though there were differences in the types of species encountered. While reptile numbers were much lower in both the wet and dry seasons, field surveys resulted in two additional species (Northern Shovel-nosed Snake and D. tinetd) being observed in the wet season. Thus, the build-up provides the best opportunity to see nocturnal species of reptiles, but season influences the type of species that may be encountered. As an additional example, the high densities of Arafura File Snakes A . arajurae found in Anniversary Creek during the early dry may be predicted following the heavy rains of the wet season because this species is aquatic and inhabits freshwater streams wherever monsoonal floods occur (Cogger 2000). Based solely on the one year of survey data, we were unable to statistically correlate the climatic conditions of the seasons with the number of reptiles; however there was a clear trend of the greatest numbers of reptiles and species being found in warm, humid seasons and the lowest numbers in the coolest, driest season. Consideration of the physiological requirements of reptiles may provide some explanation of this trend. Reptilian physiology is temperature dependent, and warmer temperatures allow reptiles to metabolise food more quickly for nutrient and water acquisition (Bennett 1982, Andrews & Pough 1985). Many species of reptile limit their activity during the dry season to balance energy' requirements and water relationships during less favourable climatic conditions (Christian & Green 1994, Christian & Griffiths 1996, Christian et al. 1999). Therefore, the high night-time temperatures and humidity of the wet, build-up and early' dry would seem ideal for nocturnal reptilian activity', and the cooler, drier conditions of the dry' would be expected to decrease activity and consequently the number of animals observed. In this study, the trend in the numbers of reptiles observed across the seasons supports expectations based on the physiological requirements of reptiles, except during the wet season. 28 Northern Territory Naturalist 18 (2005) _ L. McArthur & J. Young Table 3. Nocturnal reptile species expected to be found in the Robin Falls area (Wilson & Swan 2003, Biological Records Scheme, PWCNT), * introduced species; *** recorded in this study; SCT species susceptible to the Cane Toad (van Dam et al. 2002) indicated with an 'x'. Scientific name Recorded this study Common name SCT Crocodylidae Crocodylus johnstoni *** Freshwater Crocodile X Crocodylus porosus Saltwater Crocodile Chelidae Chelodina rugosa Northern Long-necked Turtle X Elseya dentata Northern Snapping Turtle Gekkonidae Diplodactylus conspicillatus Fat-tailed Gecko Diplodactylus stenodactylus Sand-plain Gecko Gehyra australis *** Northern Dtella Gehyra nana Spotted Rock Dtella Hemidactylus frenatus * Asian House Gecko Heteronotia binoei *** Bynoe's Gecko Heteronotia planiceps Northern Cave Gecko Nephrurus sheai Northern Knob-tailed Gecko Oedura marmorata *** Marbled Velvet Gecko Oedura rhombifer Zigzag Velvet Gecko Strophurus ciliaris cHiaris *** Northern Spiny-tailed Gecko Pygopodidae Delma borea Northern Snake-lizard Delma tincta *** Black-necked Legless Lizard Lialis burtonis *** Burton's Snake-lizard Pygopus steelescotti Northern Hooded Scaly-foot Pygopus nigriceps Black-headed Scaly-foot Typhlopidae Ramphotyphlops diversus Northern Blind Snake Ramphotyphlops guentheri Top End Blind Snake Ramphotyphlops ligatus Robust Blind Snake Ramphotyphlops tovelli Darwin Blind Snake Ramphotyphlops unguirostris Claw-snouted Blind Snake Pythonldae Antaresia childreni *** Children's Python X Aspidites melanocephalus Black-headed Python X Liasis mackloti *** Water Python X Liasis olivaceus *** Olive Python X Morelia spilota *** Carpet Python X Acrochordidae Acrochordus arafurae *** Arafura File Snake Robin Falls nocturnal reptiles Northern Territory Naturalist 18 (2005) 29 Table 3 continued Recorded Scientific name this study Common name SCT Colubridae Boiga irregularis Enhydris polyiepis Stegonotus cucullatus Tropidonophis mairii Elapidae Acanthophis praelongus Brachyurhophis roperi Rhinoplocephalus pallidiceps Furina ornata Pseudechis australis Pseudonaja nuchalis Suta punctata Vermicella intermedia Vermicella multifasciata Brown Tree Snake x Macleay's Water Snake x Slaty-grey Snake Keelback Snake Northern Death Adder x Northern Shovel-nosed Snake Northern Small-eyed Snake x Orange-naped Snake King Brown Snake x Western Brown Snake x Little Spotted Snake Wide-banded Bandy-bandy Narrow-banded Northern Bandy-bandy The significantly higher rainfall during the wet season is an additional factor to consider when attempting to understand the observed trends in reptile numbers associated with climatic variables. The quantity of rain itself is not likely to be the key, but rather the amenable environmental conditions created by excess surface water. The conundrum from this study is that fewer reptiles were observed during the wet season when it would be expected that food resources, such as frogs, would be plentiful. A study on the movements of Water Pythons, Keelbacks and Slaty-grey Snakes at Fogg Dam near Darwin had similar findings in which adults of these species were encountered less often during the wet season than in the dry season (Brown et al. 2002). The low density of adult Water Pythons encountered was explained by the migration of their prey. Dusky Rats R attus col/etti, which disperse over the floodplain during the wet season. Similarly, in this study, the Northern Death Adder Acanthophis praelongus were most commonly encountered along the Robin Falls walking track during the build-up when frog densities were at their highest along the creek, but were not encountered during the wet season when frog densities were at their lowest (McArthur & Young unpubl.). For frogs, it is assumed that the creek provides a permanent water refuge during the dry' and build-up, and that the floodplain provides suitable breeding areas during the wet. Therefore, we suspect that Northern Death Adders disperse from the creek to the floodplains during the wet to follow their prey (frogs), as do water pythons (Brown et al. 2002). Thus, while counts of reptiles were low during the wet season, it may not be indicative of inactivity but instead an artifact resulting from the seasonal dynamics of the floodplain. 30 Northern Territory Naturalist 18 (2005) L. McArthur & J. Young The advantage of choosing two different transects to conduct counts (road and walking track) is that the diversity of habitats sampled is increased and thus the diversity of reptile species encountered is similarly increased. The Brown Tree Snake (or more apdy named the ‘Night Tiger’) Boiga irregularis , and the Carpet Python Morelia spilota , were only encountered on the Robin Falls transect. The methods used for each transect were different and, therefore, any analysis direcdy comparing the number of reptiles encountered per distance between transects would be misleading. Nonetheless, it is worth noting that during the build-up the Robin Falls transect provided more individuals per distance than the Dorat Road transect (2.6 reptiles/km versus 0.22 reptiles/km, respectively). The difference in habitats (inland monsoon forest with permanent water versus dry floodplain) will likely influence the diversity of reptiles encountered. Currendy, the Cane Toad at all life stages is a concern, either directly or indirecdy, for native wildlife in the Top End (Covacevich & Archer 1975, Lawler & Hero 1997, Catling et at. 1999, Crossland 2000, van Dam et al, 2002, Altman et at. 2003). Research on long-term effects of the Cane Toad is lacking; however, based on the knowledge of diet preferences, nearly one third (30%) of the nocturnal species of reptiles local to the Robin Falls area may be susceptible to Cane Toad toxins (Covacevich & Archer 1975, van Dam et al. 2002, Wilson and Swan 2003; see also Phillips et al. 2003). Some reptile species are generalists, such as the Freshwater Crocodile, and feed on a variety of prey items (including anurans). One study has demonstrated a decline in freshwater crocodile numbers in a ‘toad invasion zone’ area relative to a ‘toad free’ area, but the same study also demonstrated persistence of the species within areas where the Cane Toad has a longer history of habitation (Catling et al 1999). Freshwater crocodiles may survive the invasion of the toad given that they either naturally avoid or 'learn' not to select toads as food items. Other species, such as the Northern Death Adder, and the Children’s Python Antaresia childrens are considered to be specialists because they feed mainly on frogs, and are more likely to be impacted by the invasion of the Cane Toad. One study has found that the Northern Death Adder need only mouth a Cane Toad to result in death (Covacevich & Archer 1975). Only two species of reptile are known to be resistant to Cane Toad toxins, the Keelback Snake, and the Slaty-grey Snake (Covacevich & Archer 1975, Phillips et al. 2003). The Keelback Snakeis one of only a few native predators able to survive Bufo toxins at all life stages: eggs. Larvae, metamorphs and adults (Ingram & Covacevich 1990, Altman et al. 2003, Wilson & Swan 2003). The tolerance of the Keelback Snake to the Cane Toad is 'attributed to the long evolutionary association of natricines and bufonids’ in areas such as North and South America, Asia and Europe, which links it with close relatives in the subfamily Natricinae (Ingram & Covacevich 1990). The future of the other reptiles listed in this study, that are not considered to be susceptible to Cane Toad toxins, will be at less risk because these species specialise on different prey items. For example, the Northern Shovel-nosed Snake consumes only reptile eggs; the Orange-naped Snake Furiua oniata, eats exclusively skinks; and the two Robin Falls nocturnal reptiles Northern Territory Naturalist 18 (2005) 31 bandy-bandys, I 'ermicella spp., feed only on blind snakes (Ramphotyphlops spp.) (Cogger 2000 ). This study has been useful in providing a list of nocturnal reptiles of the Robin Falls area, which denotes in which season animals were found. Twenty-two species of nocturnal reptile were identified and a total of 90 individuals were observed during 36 survey sessions. While this survey of the Robin Falls area provides very preliminary data on the types of nocturnal reptiles and their numbers, it also provides quantitative information that may prove useful given the imminent arrival of the Cane Toad. Acknowledgements Many thanks go to Keith Christian’s laboratory, Charles Darwin University, for supplying resources and support to conduct this research. And man}' thanks to the following people for commenting on the manuscript: Gavin Bedford, Paul Horner, )ames Smith and Sebastian Iglesias. The following volunteers are most gratefully appreciated for their efforts and late hours spent helping with this research: Carola Bednarik, Jenny Brazier, Geoffrey Carr, Peter Clifton, Kate Goodrich, Birgit Hallenstein, Carolyn Haynes, Sommer Jenkins, Jenny Kapp, Sally Kent, David Rose, James Smith, Victor Vilar, Martin Whiting. References Altman J., Griffiths T. and Whitehead P. (2003) Invasion of the nibbish frogs. Nature Australia 27(10), 44-51. Andrews R.M. and Pough H.F. (1985) Metabolism of squamate reptiles: allometric and ecological relationships. Physiological Zoology 58, 214-231. Bennett A.F. (1982) The energetics of reptilian activity. In Biology ofRepti/ia (ed. C. Gans & F.H. Pough), pp. 155-199. Academic Press, London. Brown G.P., Shine R. and Madsen T. (2002) Responses of three sympatric snake species to tropical seasonality in northern AusmYia. Journal of Tropical Eco/ojy 18, 549-568. Catling P.C., Hertog A., Burt R.J., Wombey J.C. and Forrester R.l. (1999) The short-term effect of cane toads ( B/ifo marinas) on native fauna in the Gulf Country of the Northern Territory'. Wildlife Research 26, 161-185. Christian K., Bedford G., Green B., Griffiths A., Newgrain K. and Schultz T. (1999) Physiological ecology of a tropical dragon, I jsphogiathns temporalis. Australian Journal of Eco/ojy 24, 171-181. Christian K.A. and Green B. (1994) Seasonal energetics and water turnover of the Frillneck Lizard, Cblantydosaurus kingii, in the wet-dry tropics of Australia. Herpetohsjca 50,274-281. Christian K.A. and Griffiths A.D. (1996) Physiological ecology of frillneck lizards in a seasonal tropical environment. Oeco/ogia 106, 49-56. Cogger H.G. (2000) Reptiles and Amphibians of Australia. Reed New Holland, Sydney. Covacevich J. and Archer M. (1975) The distribution of the cane toad, Bnfo marinas, in Australia and its effects on indigenous vertebrates. Memoirs of the Queensland Museum 17, 305-310. Crossland M.R. (2000) Direct and indirect effects of the introduced toad Bnfo marinus (Anura: Bufonidae) on populations of native anuran larvae in Australia. Bcography 23, 283-290. 32 Northern Territory Naturalist 18 (2005) L. McArthur & J. Young Crump M.L. and Scott N.J. (1994) Visual encounter surveys. In Measuring and Monitoring Biological Diversity. Standard Methodsfor .Amphibians (ed. YV.R. Heyer, M.A. Donnelly, RAV. McDiarmid, L.C. Hayek and M.S. Foster), pp. 84-91. Smithsonian Institution Press, W ashington. Edstrom A. (1992) Venomous and Poisonous Animals. Krieger Publishing Company, Florida. Frogwatch (2004). http: / /www.fropwatch.org.au /canetoads /index.cfm . Ingram G.J. and Covacevich J. (1990) 7 ropidonophis mairii vs Bnfo marinas. Memoirs of the Queensland Museum 29, 396. Lawler K.L. and Hero J.-M. (1997) Payability of Bnfo marinas tadpoles to a predator}' fish decreases with development. Wildlife Research 24, 327-334. List R.J. (1971) Smithsonian Meteorological Tables Publication 4014. Smithsonian Institution, Washington D.C. Phillips B.L., Brown G.P., and Shine R. (2003) Assessing the potential impact of cane toads on Australian snakes. Conservation Biology 17, 1738-1747. Straughan I.R. (1966) The natural history of the "Cane Toad" in Queensland. Australian Natural History 15, 230-232. Sutherland S.K. (1983) Australian . Animal Toxins: The creatures, their toxins and can of the poisoned patient. Oxford University Press, Oxford. van Dam R.A., Walden D.J. and Begg G.W, (2002) A preliminary risk assessment of cane toads in Kakadu National Park. Supervising Scientist Report 164,1-108. Wilson S. and Swan G. (2003) A Complete Guide to Reptiles of Australia. Reed New Holland, Sydney. The Northern Spiny-tailed Gecko Strophurus dHaris ciHaris was recorded near Robin Falls mosdy in the dry season. (Lorrae McArthur) Robin Falls nocturnal reptiles Northern Territory Naturalist 18 (2005) 33 The resemblance of Burton's Snake-lizard Iialis burtonis to snakes is superficial. (Lorrae McArthur) The Wide-banded Bandy-bandy Vermicelli! intermedia feeds exclusively on blind snakes and may thus avoid the direct consequences of Cane Toad toxicity. (Lorrae McArthur) Northern Territory Naturalist (2005) 18: 34-44 Distribution and natural history of the cryptic Chameleon Dragon Chelosania brunnea: a review of records Colin R. Trainor Tropical Savannas Management Cooperative Research Centre, Charles Darwin University, NT 0909. Email: colin.trainor@cdu.edu.au Abstract The Chameleon Dragon Chelosania brunnea is a rarely encountered and poorly-known agamid lizard. A total of 103 records (58 specimen-backed) were collated to examine aspects of its biolog)'. The Chameleon Dragon occurs widely across north and north¬ western Australia with the majority’ of records from Kakadu National Park (41% of all records), the Dampier Peninsula, Kimberley and Katherine areas, largely reflecting past wildlife survey effort. Climatic modelling using BIOCLIM predicted its occurrence in extensive areas of the Gulf of Carpentaria and southern Cape York. The Chameleon Dragon is arboreal, apparently' sedentary-, with almost all records from eucalypt forest and woodland. Breeding is highly seasonal with males observed mostly in May (apparently to select mates) and females recorded mostly in the mid-dry season (July-August) as they descend to dig nest holes. A juvenile has been recorded in January'. Hot late dry season wildfires are probably the main threat to Chameleon Dragons. Suggestions are given for further study of this interesting species. Introduction The monotypic Chameleon Dragon Chelosania brunnea (Gray' 1845) is a medium-sized (snout-vent 105 mm), large-headed, short-limbed agamid lizard that is restricted to the wet-dry' tropical forests and woodlands of northern Australia (Cogger 1994). The Chameleon Dragon is named because of its superficial similarity to Chameleon Dragons (Family Chameleonidae), to which it is unrelated; a spectacular example of independent covergent evolution, just two short notes, totalling fewer than 500 words, have been published on the species. Greer (1990) described die Chameleon Dragon as a ‘peculiar beast about which little is known’. Undl the 1960s, the distribution of the Chameleon Dragon was reported as Western Australia (Worrell 1963), however this ignored museum specimen records from near Darwin and Arnhem Land, collected in the early 20 th century. One of the first published field reports was in 1979, with notes on a significant range extension and behaviour (Husband 1979). After releasing a Chameleon Dragon, Husband (1979) Chameleon Dragon records Northern Territory Naturalist 18 (2005) 35 noted that ‘when moving through the branches it was very chameleon-like, slowly almost mechanically moving from point to point,..’. A second publication described basic details of reproduction (Pengiliey 1982). Miles and Burbidge (1975) and Bush (1985) summarise several additional Western Australian observations. This paper collates known records of the Chameleon Dragon (to August 2004), develops a predicted distribution by correlative bioclimatic analysis (BIOCL1M) and documents biological details (habitat use, breeding and seasonality) gathered from specimen and unpublished anecdotal records of biologists and field naturalists. Methods Data sources Chameleon Dragon locality records were sought from Australian and international museums, from biological adas/database schemes managed by Australian state and territory natural resource management departments, and by seeking published and unpublished site records from field biologists and herpetologists (all are listed in acknowledgements or cited in the text). Data sought was locality, georeference, date, habitat and behaviour. Records with sufficient information on habitat were subjectively classified into five broad habitat types: eucalypt forest, eucalypt woodland, tropical dry forest (= monsoon rainforest), acacia woodland and urban gardens. Predicting distribution Known geo-referenced locality records (« = 80), except for the 23 records obtained after a June 1997 cut-off date, were entered into BIOCL1M to predict the distribution of the Chameleon Dragon. BIOCLIM is a bio-climatic analysis and prediction system, which can generate site-specific estimates of monthly temperatures and precipitation for sites anywhere in Australia. Results of this analysis are noted briefly. Results and discussion Known distribution The Chameleon Dragon is widely distributed in Australia’s wet-dry tropics across almost six degrees of latitude (12°—17°30S) and 16 degrees of longitude (122"—138" E): a land area of c. 800,000 km 2 . The known distribution of the Chameleon Dragon is given in Figure 1. The BIOCLIM-predicted distribution encompassed known records but furthermore predicted occurrence in the Qld Gulf of Carpentaria, where the Chameleon Dragon has never been recorded (but which has been litde surveyed). 36 Northern Territory Naturalist 18 (2005) C. Trainor Figure 1. Distribution of the Chameleon Dragon. Location numbers are listed in the Appendix. The scale bar relates to the main map. The inset map is Kakadu National Park in the Northern Territory. Western Australia In WA the majority of records are specimen-based, with records from the Dampier Peninsula, Derby and Leopold Ranges area. Storr et al. (1983) list far north Western Australia including the Kimberley to Coulomb Point, King Leopold Range and Kununurra in its range. 'ITtere are several records throughout the north Kimberley, Drysdale River National Park and Kalumburu area, but few from the central Kimberley Plateau, probably reflecting the lack of survey effort and inaccessibility of this area. Residents of Kununurra reported several observations of Chameleon Dragon within the town over the past 25 years (particularly at caravan parks), and there are two records from Bungle Bungle National Park (including a specimen). Northern Territory In the NT there is one record of this dragon from the far west, from Spirit Hills Station near Kununurra, with a major gap in distribution records between the Victoria-Bonaparte and Daly Basin districts (probably reflecting low survey effort in Chameleon Dragon records Northern Territory Naturalist 18 (2005) 37 this remote area). Similarly there is only one historical record of Chameleon Dragon for the Darwin area. Many biologists are based in Darwin, so this may represent a real range gap. Almost half of all known records are from Kakadu National Park (KNP) (particularly Kapalga Research Station, Jabiru, Nourlangie and Ubirr Rock) and Oenpelli in western Arnhem Land, highlighting the high field survey effort. There are several records for the Katherine and Mataranka districts, with only two in the Daly Waters-Sturt Plateau area (including Husband 1979). There are two (or three) records from the far west of the NT Gulf Region on Wollogorang Station. The pattern of records for die NT appears to reflect previous survey effort, but the increased frequency of records from the higher rainfall forests may also relate to a habitat preference. Queensland The Chameleon Dragon was included on the official wildlife list for Qld in 1991 (Covacevich and Couper 1991) because of‘an unambiguous photo’ of a Chameleon Dragon taken in 1978 at ‘Camp Ridgeway’, between the NT/Qld border and the Westmoreland Station homestead by a geologist (H. Cogger, pers. comm.). It was subsequendy removed from the list because no specimens were available (P. Couper, pers. comm.). Habitat use The Chameleon Dragon prefers eucalypt forests and woodlands, with few records from dry tropical forest (monsoon rainforest), fioodplain, acacia woodland and urban areas (Cogger 1981; this review). There were two records from tropical dry forest (Figure 2) including a Wollogorang record where an individual was collected from a small patch of forest. Apart from the record by Husband (1979) from acacia woodland (presumably Lancewood Acacia sbirleyi scrub), there was a single specimen collected from the same habitat near Kalumburu by the Australian Museum. One naturalist observed a Chameleon Dragon sleeping in ‘Pindan’ acacia trees north of Broome and suspects that they might be associated with acacia shrubland on deep sand (I. Morris, pers. comm.) but none were recorded during biological surveys targeting such shrubland (Woinarski tk Fisher 1995a,b). There are two reports from the vicinity of towns: at Kununurra and Oenpelli, and several more undocumented reports from Jabiru (G. Miles, pers.comm.) (see Appendix). Behaviour and microhabitat use Chameleon Dragons are mostly arboreal and sedentary. Individuals are repeatedly observed at regular points along roads in KNP indicating that they occupy permanent territories (I. Morris, pers. comm.). A total of 10 individuals have been captured in pitfall traps in KNP (T. Hertog, pers. comm) and at Kalumburu (J. Wombey, pers. comm.). Trap capture rates are exceptionally low as the KNP data represent c. 4,000 pitfall trap nights over a 7 year period. Chameleon Dragons are most often observed 38 Northern Territory Naturalist 18 (2005) C. Trainor on the ground (on tracks, as road-killed individuals, basking, fighting or digging burrows), six have been observed low in trees and none have been recorded from the canopy (Table 1) although it is suspected that they spend most time at this level. They are probably experts at evasion, as they ‘slip around the tree to avoid predators’ (N. Gambold, pers. comm.). They are also inexplicably clumsy. One individual landed on the windscreen of a moving car, apparently after being captured and then released by a diurnal raptor (Museum and Art Gallery of the Northern Territory database). A Chameleon Dragon fell from the canopy of Eucalyptus woodland in Kakadu (A. Dudley, pers. comm.), and at Wollogorang a dragon fell from the canopy of eucalvpt woodland onto a tent (C. Trainor, pers. obs.). Tony Griffiths (pers. comm.) observed a released animal climb, then fall off the lower trunk of a tree. Little is known of diet. In Kakadu an adult male road-killed on 5 July 2002 had: “Stomach quite full; contained about 80 green ants [Green Ants Oecopbylla maragdina ], plus 4-5 square bits of bark. There was a fecal pellet in intestine, also wholly green ants” (S. Sweet, pers. comm.). A juvenile dragon has been observed feeding on Green Ants (I. Morris, pers. comm.). A ranger observed a juvenile sleeping/basking on a tree root in KNP (L. Barnett per I. Morris, pers. comm.). After release, a male animal at Kapalga slowly moved 2 m then climbed a Eucalyptus tetrodonta tree, moving its limbs slowly with little use of its hind feet while climbing. The tail is prehensile, and during this observation it was used to grip branches and other objects (T. Griffiths, pers. comm.). eucalypt forest eucalypt woodland dry tropical forest acacia woodland urban gardens 0 5 10 15 20 25 30 No. of records Figure 2. Frequency of occurrence of the Chameleon Dragon in broad habitat types. Chameleon Dragon records Northern Territory Naturalist 18 (2005) 39 Table 1. Frequency of occurrence of microhabitat use (including pitfall trap) by Chameleon Dragon when first sighted (« = 41 records). * = includes two records of an animal that fell to ground from the forest canopy, and one that was dropped to ground by a diurnal raptor. Microhabitat Frequency Pitfall trap capture 10 On low tree (<2 m) 6 On ground: stationary or crossing track 10 On ground 5 * On ground: road kill 4 On ground: digging burrow 2 On ground: territorial fight 2 On ground: with predator 1 On ground: basking 1 Breeding, seasonality and detectability Half of the 75 dated locality records are from July, August and September (Figure 3). Males are usually observed in May (5 of 9 records where sex known), and females in July and August (9 of 12 records where sex known). Presumably males descend at this time to take part in territorial fighting to establish mating rights (see below). In the following months, gravid females descend to the ground to lay eggs, with several direct observations of this behaviour (|. Wombey, pers. comm.). A juvenile dragon (snout-vent 53 mm, weight 3 g) was captured on 19 January 1979 (N. Sonnemann, pers. comm.), probably having hatched in the late dry season (September-October). Basking by gravid females may increase detectability because they are heavier and less mobile than non-gravid individuals (G. Miles, pers. comm.). An adult female, suspected of being gravid, ‘walked in a slow, rather jerky gait’ after being released (J. Wombey, pers. comm.). Females are known to lay eight eggs (G. Miles, pers. comm.). The egg number, incubation period, egg and hatchling morphometries of a purported single clutch was described by Pengilley (1982), but he had described the characteristics of two clutches rather than one (G. Miles, pers. comm.). It has been speculated that in the wet season they ‘prefer sandy substrates which are boggy and impassable in the wet season’, which may further reduce opportunities to observe dragons in this season (G. Miles, pers. comm.). The Chameleon Dragon may have 40 Northern Territory Naturalist 18 (2005) C. Trainor been observed less frequently in KNP since roads were upgraded to bitumen (c. 1985), because the black surfaces are too hot for basking and visual detection is reduced (I. Morris, pers. comm.). Figure 3. Number of monthly records of the Chameleon Dragon (« — 75 dated records). Relative abundance Little is known of the Chameleon Dragon’s relative abundance, but field observations are infrequent. Several experts consider that they are not rare, but difficult to record unless gravid (P. Harlow and J. Wombey, pers. comm.) or involved in territorial fights. The number of records has increased greatly over the last three decades of the 20th century', with the intensification of biological field survey in northern Australia. Major fauna surveys have been notably unsuccessful at recording Chameleon Dragons: one was recorded during surveys of KNP Stages 1 and 2 over about 3 person years (Braithwaite 1985); none were recorded in the KNP Stage 3 survey, involving about 200 person days (Woinarski & Braithwaite 1991), or during a Bungle Bungle National Park survey (Woinarski 1992). None were recorded during surveys of Litchfield, Elsey or Limmen Gate Nat. Parks (Griffiths 1997, Griffiths et at. 1997ab) or the continental Wessel and English Company Islands (Woinarski et ai 1999). Chameleon Dragon records Northern Territory Naturalist 18 (2005) 41 There are two confirmed observations of more than one individual. On 1 May 1979, two males were observed in a territorial fight in KNP, and then a female was observed nearby on the same day (J. Wombey, pets. comm.). At Kapalga, during June 1992, two males were observed by the author on the ground clasping each other’s bodies while presumably fighting. Additionally, a total of 10 individuals were collected from the Wotjulum area of VVA in 1954-1955 during museum surveys, which comprise almost all the records for that decade. Four individuals were observed in one day by a ‘Mr Bishop’ (N. Gambold, pers. comm.). At Kununurra four Chameleon Dragons were observed over 10 months in 1978-1979 (N. Sonnemann, pers. comm.) and in the Jabiru area of KNP, about eight Chameleon Dragons were collected over a 22 yr period (G. Miles, pers. comm.). Possible threats Hot late season fires which burn into the canopy would be the greatest threat to Chameleon Dragon populations, as most individuals would be killed by this regime. Limited data from the Kapalga Fire Experiment provides some support for this opinion. During reptile sampling, only seven Chameleon Dragons were captured over five years. Four were recorded in control years before the burning began (two in progressive burn samples and two in late bum samples). After the experiment began, Chameleon Dragons were recorded only in unburnt or early burn sites (T. Hertog, pers. comm.; the author, pers. obs.). Predation by Feral Cats Felts catus, has been suggested as threats to dragons (N. Gambold, pers. comm.; B. Hancock, pers. comm.). A cat was observed mauling a Chameleon Dragon at Kununurra (N. Sonnemann, pers. comm.). Chemical spraying (with dieldrin) has also been suggested as a cause of the reduced frequency of observations of dragons in the vicinity of the town of Jabiru (I. Morris, pers. comm.). Conservation and suggestions for further study Although rarely recorded, the Chameleon Dragon is not considered globally threatened (http//www.red 1 ist.org, 2005) or regionally rare or threatened, however Gambold and Menkhorst (1992) used the presence of Chameleon Dragons as supporting evidence for a National Estate proposal. More than half of the 103 records are from protected areas (’Fable 2), reflecting survey effort in designated protected areas and presumably demonstrating that the network provides good coverage of dragon habitat. There are at least 58 museum specimens (see Appendix) that could be analysed for study of diet, reproduction and genetics. Field studies will need to focus on active searching during the early and mid-dry season when males and females descend to the ground to mate and lay eggs. Pitfall trapping has rarely been successful. The Kununurra area, particularly around caravan parks, KNP area, Kalumburu and Cape Leveque may be suitable locations for field study. The Bardi people (at least) of the Dampier Peninsula are reputedly familiar with the Chameleon Dragon (I. Morris, pers. 42 Northern Territory Naturalist 18 (2005) C. Trainor comm.). Questionnaires and interview with Aboriginals over the range of the species, aiming to document local knowledge of occurrence, habitat use and behaviour also has potential to reveal new ecological details of this elusive dragon lizard. Table 2. Frequency of Chameleon Dragon records (« — 60) from protected areas and Aboriginal Land Trusts in north-western Australia. Protected areas Area (km 2 ) No. records Kakadu National Park (KNP) 20,000 42 Umbrawarra Gorge Nature Park 20 1 Point Coulomb Nature Reserve c. 200 1 Drysdale River National Park 7,000 3 Bungle Bungle National Park 3,100 2 Cutta Cutta Caves Nature Park 20 2 Arnhem Land Aboriginal Land Trust c. 60,000 8 Prince Regent River Nature Reserve 6,300 1 Acknowledgements Thanks to James Smith of Charles Darwin University who experdy prepared the map. The following individuals kindly sought and provided dragon records: Gavin Bedford (Charles Darwin University, Darwin), Kym Brennan (Paries and Wildlife Commission of the Northern Territory, Winnellie), Brian Bush (Stoneville, WA), Ray Chatto (Parks and Wildlife Commission of the Northern Territory, Winnellie), Hal Cogger (Australian Museum/Pearl Beach), Patrick Couper (Queensland Museum, Brisbane), Ronald Crombie (Smithsonian Institute, Washington), Alex Dudley (Warialda, NSW), Gordon Graham (Department of Conservation and Land Management, WA), Tony Griffiths (Charles Darwin University, Darwin), Bryan Hancock (Kununurra, WA), Peter Harlow (University of Sydney), Tony Hertog (Commonwealth Scientific and Industrial Research Organisation, Winnellie), Paul Horner (Museum and Art Gallery of the Northern Territory, Darwin), Grant Husband (Territory Wildlife Park, Parks and Wildlife Commission of the Northern Territory'), Ivan Ineich (Museum Nationale D’Historie Naturelle, Paris), Chris Johnson (Parks and Wildlife Commission of the Northern Territory, Winnellie, now Ireland), Dave Lindner (Department of Environment and Heritage, Jabiru), Gunther Kohler (Scnckenburg Forschungininstitut Naturmuseum, Frankfurt), Colin McCarthy (Natural History Chameleon Dragon records Northern Territory Naturalist 18 (2005) 43 Museum, London), Keith McDonald (QLD), Greg Miles (Department of Environment and Heritage, Jabiru), Nic Gambold (Central Land Council, Alice Springs), Ian Morris (Noonamah, NT), Cassy Redhed (Chicago Field Museum), Ross Sadlier (Australian Museum, Canberra), M. Ruf (Zoological Museum of Zurich), Rick Shine (University of Sydney), Martin Shultz (Southern Cross University), Neil Sonnemann, Laurie Smith (Western Australian Musem, Perth), Sam Sweet (University of California), Steve Wilson (Queensland Museum, Brisbane) and John Wombey (Australian National Wildlife Collection, CSIRO, Canberra). References Braithwaite R.W. (ed.) (1985) The Kakadu Fauna Survey: an ecological survey of Kakadu National Park. Report to the Australian National Parks and Wildlife Sendee, from CSIRO, Wildlife and Ecology, Danvin. Bush B. (1985) On the distribution of two north Australian agamid lizards, Cbelosania brunnea Gray and Cblanrydosaurus kingti Gray. West Australian Naturalist 16 (2/3), 53. Cogger H.G. (1981) A biogeographic study of the Arnhem Land herpetofauna. Proceedings of the Melbourne Heipeto/ogicalSymposium 148-155. Cogger H.G. (1994) Reptiles and Amphibians of Australia. Reed Books, Chatswood. Covacevich J.A and Couper P.J. (1991). The Reptile Records. In An Atlas of Queensland! Frogs, Reptiles, birds and Mammals) .(eds Ingram G. J. and Raven R. J.). Pp 45-140. Queensland Museum, Brisbane. Gambold N.J. and Menkhorst K.A. (1992). Katherine Region Heritage Study: significant naturalfeatures and the impact of Tourism. Report to the Australian Heritage Commission from the Environment Centre Inc., Danvin. Greer A. (1990) The biology and Evolution of Australian I j\ards. Surrey Beatty and Sons, Chipping Norton. Griffiths A.D. (1997) biological survey of EJsey National Park. Technical Report No. 63, Parks and Wildlife Commission of the Northern Territory, Darwin. Griffiths A.D., Woinarski J.C.Z., Armstrong M.D., Cowie IX)., Dunlop C.R.and Horner P.G. (1997a) biological survey of Utchfield National Park. Technical Report No. 62, Parks and Wildlife Commission of the Northern Territory, Danvin. Griffiths A.D., Materne C.M. and Shenvell D.J. (1997b) biological survey of the proposed Limmen Cate National Park. Technical Report No. 61, Parks and W'ildlife Commission of the Northern Territory, Danvin. Husband G. (1979) Range extension for Cbelosania brunnea. Herpetofauna 10, 29-30. Miles J.M and Burbidge A.A. (eds) (1975) A biological survey of the Prince Regent Reserve North-West Kimberly, Western Australia in August, 1974. Department of Fisheries and Wildlife, Wildlife Research Bulletin No. 3, Perth. Pengilley R. (1982). Note on the reproductive biology of the Ring-tailed Dragon ( Cbelosania brunnea). Northern 'Territory Naturalist 5, 6. Storr G.M., Smith L.A. and Johnstone R.E. (1983) IJ^ards of Western Australia II. Dragons and monitors. Western Australian Museum, Perth. Woinarski J.C.Z. (ed.) (1992) The nildlife and vegetation ofPumulutu (bungle bungle) National Park and adjacent area. Department of Consenration and Land Management, Wildlife Research Bulletin No. 6, Perth. 44 Northern Territory Naturalist 18 (2005) C. Trainor Woinarski J.C.Z. and Braithwaite R.W. (1991) Wildlife of Kakadu Stage III, a synthesis: report to Australian National Parks & Wildlife Service. Winnellie: CSIRO Tropical Ecosystem Research Centre. Woinarski J.C.Z. and Fisher A. (1995a) Wildlife of Lancewood {Acacia sbirleyi) thickets and woodlands in Northern Australia. 1. Variation in vertebrate species composition across the environmental range occupied by Lancewood vegetation in Northern Territory. Wildlife Research 22, 379-411. Woinarski J.C.Z. and Fisher A. (1995/t) Wildlife of Lancewood {Acacia sbirleyi) thickets and woodlands in Northern Australia. II. Comparisons with other environments of the region {Acacia woodlands. Eucalyptus savanna woodlands and monsoon rainforests). Wildlife R esearch 22,413-443. Woinarski J.C.Z., Horner P., Fisher A., Brennan K., Lindner G., Gambold N., Chatto R. and Morris I. (1999) Distributional patterning of terrestrial herpetofauna on the Wessel and English Company Island groups, northeastern Arnhem Land, Northern Territory, Australia. Australian Journal of Ecology 24, 60-79. Worrell E. (1963). Reptiles of Australia; Crocodiles, Turtles, Tortoises, I awards, Snakes. Angus and Robinson, Sydney. Appendix. Chameleon Dragon locality record details. This appendix may be viewed at: http:/ / www.geocities.com/ntfieldnaturalists/journal/Trainor appendix.pdf The Chameleon Dragon Chelosania brutmea is arboreal, although females descend to the ground to dig nest holes. (Martin Armstrong) Northern Territory Naturalist (2005) 18: 45-50 Avian granivores consume flowers, not just seed, of the Top End Bamboo Bambusa arnhemica Donald C. Franklin School for Environmental Research, Charles Darwin University, Darwin, NT 0909. Email: don.franklin@cdu.edu.au "Who eats bamboo seed? Everybody does." So wrote Janzen (1976) in his classic work on the gregarious flowering and subsequent die-off of many bamboos. Janzen collated a large anecdotal record demonstrating aggregation (often in huge numbers) at flowering stands of bamboo, and consumption of bamboo seed, by humans, many species of rodent, pigs, elephants, rhinoceros, a range of bovids and cervids, monkeys, a variety of birds, and insects. During a drought in India, an estimated 35 000 people survived on bamboo seed (Lowrie 1900 cited in Janzen 1976). Mass-flowering events are often followed by rodent plagues (Chauhan & Sasena 1985, Jaksic & Lima 2003). At least one bird species, the Pied Mannikin Lonchura fringil/oides of East Africa, appears to be a bamboo seed specialist (Jackson 1972). The seeds of most bamboos are large, nutritious grains (caryopses) that lack defensive toxins (Pathak 1979, Bhargava cl al. 1996). Janzen (1976) proposed that the peculiarly infrequent gregarious-flowering behaviour and subsequent death of certain bamboos was an evolutionary response to the vulnerability of its seeds to consumers, in which huge production of seed alternates with many years of no production at all, the latter ensuring that populations of seed consumers cannot build up to match the abundance of the resource - the seed predator satiation hypothesis. Effective seed predator satiation during a mast-fruiting event, and subsequent total failure of seed under unsatiated conditions, has been documented in Bornean dipterocarps (Curran & Webb 2000). However, the improved pollination rates achievable by wind-pollinated species when neighbours are flowering, along with several other possibilities, are potential additional or alternative explanations for the adaptive value of infrequent but massive fruiting events (Kelly 1994, Sakai 2002). Bambusa arnhemica , a bamboo endemic to the north-western Top End of the Northern Territory 1 , flowers gregariously at intervals of c. 40—50 years and seeds prolifically (Franklin 2004). Simple or compound flowering branches up to 2.5 m long bear clusters (inflorescences) of pseudospikelets at each node. In this note, I report observations, obtained in the course of studies of B. arnhemica flowering behaviour, of birds consuming plant material from the inflorescences of B. arnhemica. When 1 located birds amongst standing B. arnhemica inflorescences, I examined them through binoculars. If any individual was observed consuming any part of the inflorescence (cf probing in apparent search of insects), 1 noted the species, and counted or estimated the flock size. To maximise independence of records, a bird species was 46 Northern Territory Naturalist 18 (2005) D. Franklin recorded in this way at a maximum of one location per day. I also report aviary trials to test my suspicions about what the birds were consuming. I obtained 93 observations of 11 bird species foraging on the inflorescences of B. amhemica, comprising four cockatoo, three parrot and four finch species (Table 1). Reflecting where I spent most time in flowering bamboo stands, 83 records (89%) were in the Adelaide River catchment, the remainder being shared between the Daly, Finniss and Mary River catchments. Reflecting the main flowering and seeding times of B. amhemica (Franklin, unpubl.), 89 (96%) of records were within the period July to December, with a peak in October (28 records, 30%). The most frequently recorded and numerous species were Rainbow Lorikeet and Chestnut-breasted Mannikin, followed by Double-barred Finch and Red-tailed Black-Cockatoo. Rainbow Lorikeets fed particularly persistently, moving slowly from inflorescence to inflorescence in a manner that could feasibly have serious adverse affects on the productivity of clumps, especially if only isolated clumps were flowering. Mannikins, on the other hand, tended to move through an area, feeding fairly briefly on each clump. Cockatoos may be relatively under-reported because they were often wary and flushed before I could confirm their behaviour. Several small honeyeater species, notably the White-throated Honeyeater Melithreptns albogularis and the Brown Honeyeater ljchmera indistincta, were frequently recorded probing inflorescences, but I saw no evidence that they were consuming plant material. Granivorous birds were also observed foraging on the ground below flowering or recently-seeded bamboo clumps, but the identity of the food being consumed could not be confirmed. Table 1. Birds observed feeding at inflorescences of wild bamboo Bambusa amhemica. Species Median flock No. of records size (range) Cacatuidae Red-tailed Black-Cockatoo Calyptorhynchus banksii 9 10 (2-100) Galah Cacatua roseicapilla 1 4 (4) Little Corella Cacatua sanguinea 4 8.5 (7-13) Sulphur-crested Cockatoo Cacatua galerita 4 1 (1-2) Psittacidae Rainbow Lorikeet Trichoglossus haematodus 21 5 (2-60) Red-winged Parrot Aprosmictus erythropterus 7 2 (1-4) Northern Rosella Ptatycercus venustus 5 2 (1-5) Passeridae Double-barred Finch Taeniopygia bichenovii 14 3 (1-10) Long-tailed Finch Poephila acuticauda 1 8 (8) Crimson Finch Neochmia phaeton 7 2 (1-5) Chestnut-breasted Mannikin Lonchura castaneothorax 20 10 (1-30) Bamboo and birds Northern Territory Naturalist 18 (2005) 47 I started with the assumption that the listed bird species were seeking and consuming bamboo seed, but soon became suspicious that this was frequently not the case. I noticed that bird activity' at the inflorescences appeared related to flowering perse and not to my unquantified observations about seed availability, the latter being sparse except in October or November. On a number of occasions when I saw much active foraging amongst the inflorescences, I made an effort to locate seed on them, but found few. Furthermore, the rapid and repetitive bill movements of foraging granivores that 1 observed repeatedly was hardly consistent with either direct consumption or fragmentation of seeds. So far as I could determine, finches and Rainbow Lorikeets appeared to be chewing 'empty 1 green florets, but I could not determine what the cockatoos were consuming. My suspicions that flowers were being consumed were further raised when I measured B. ambemica seeds. These are grains with a mean air-dry' weight of 19.3 mg and mean dimensions of 7.6 by 2.1 mm (Franklin 2003). They weigh 3.2 times as much as seeds of Annual Spear-grass Sarga intrans and 1.5 times those of Giant Spear-grass Ueteropogon tritueus (Dostine & Franklin 2002). In a study of finch diets (Gouldian Etytbrura goi/ldiae; Masked Poepbila personata; and Long-tailed P. acuticaudcr, Dostine & Franklin 2002), seeds of S. intram were the largest seed consumed whole, those of H. tritueus being consumed only after being fragmented. As Crimson Finch and especially Double-barred Finches are smaller-bodied and have smaller bills dtan any of these species (Franklin, unpubl.), and bill size in Australian grass-finches is direcdy related to the size of seed consumed (Todd et al. 2003), it seemed most unlikely that these birds could swallow whole B. ambemica seeds. To test the interest of birds in inflorescences lacking seed, and to identify what else they might be consuming, I collected sections of B. ambemica flowering branches, each around 15-20 cm long and containing two or three inflorescences, from the Adelaide River. I checked each section thoroughly and removed the few ripe or un-ripe seeds, which were readily detected by the swollen firmness of the floret. On the same afternoon, I took them to the Territory Wildlife Park, where two to three sections were placed in each of four small aviaries, either attached to the wire or on the ground (as advised by keeping staff based on what the birds were used to). Aviaries were selected to represent a range of the finches and parrots available including, where possible, the species listed in Table 1. The birds' response was observed from a distance of from 3—5 metres for about 10 minutes. Rainbow Lorikeets and Hooded Parrots showed no sign of interest in the branch sections, and after several 'exploratory'’ pecks they were ignored by Pictorella Mannikins (Table 2). However, Yellow-rumped Mannikins, Gouldian Finches and one Red-winged Parrot fed persistently and systematically on the inflorescences. The Red-winged Parrot worked over each inflorescence, breaking off most or all of an entire pseudospikelet before running it sideways through its bill and masticating it thoroughly, then dropping the remaining green material (the lemma and palea). 48 Northern Territory Naturalist 18 (2005) D. Franklin Yellow-rumped Mannikins worked along the spikelets, often removing one or more florets and masticating them, shedding the green matter afterwards. Gouldian Finches pecked persistently at the inflorescences, working along the spikelets, but did not obviously remove anything. In the case of Yellow-rumped Mannikins and the Red-winged Parrot, it was clear that the internal contents of the florets, and not the outer glumes, were sought and consumed, and that these were obtained by squeezing out the contents. It is not clear what the Gouldian Finches were doing or obtaining, but they could have been obtaining droplets or fragments left by the mannikins. The observed foraging behaviour by Yellow-rumped Mannikins and the Red-winged Parrot was consistent with what I had observed in the wild. Table 2. Response of aviary birds to provision of seedless Bambusa amhemica inflorescences, 22 August 2001. Aviary Species Response to inflorescences 1 2 Red-winged Parrots Aprosmictus erythropterus 2 Rainbow Lorikeets Trichoglossus haematodus one Red-winged Parrot systematically consumed them 2 c. 15 Pictorella Mannikins Heteromunia pectoralis initial interest only; no consumption 3 c. 15 Gouldian Finches Erythrura gouldiae 5 Yellow-rumped Mannikins Lonchura flaviprymna persistent consumption by both species, repeated when a second batch of inflorescences were provided 4 4 Hooded Parrots Psepholus dissimilis no sign of interest The contents of B. amhemica florets, comprising the ovary, lodicules, stigmas and stamens, are evidently of considerable interest as a food source to a range of birds. Whilst such foraging behaviour is not surprising in the dietarily versatile Rainbow Lorikeet and other parrots, and perhaps also amongst cockatoos (Higgins 1999), there is no precedent for it amongst Australian finches which are regarded as either strictly granivorous or seasonally somewhat insectivorous (Immelmann 1982, Read 1994, Dostine & Franklin 2002, Todd et al. 2003). Todd et al (2003) did, however, report that Crimson Finches consumed lerp, a soft carbohydrate exudate produced by psyllids. There appear to be no published studies of the diet of the main finch species involved here, the Chestnut-breasted Mannikin and Double-barred Finch, but it may be of relevance that these two species and especially the mannikin are somewhat noteworthy for the frequency with which they forage from standing grass stems rather than on the ground (Immelmann 1982), a habit which might predispose them to arboreal foraging on bamboo. Bamboo and birds Northern Territory Naturalist 18 (2005) 49 Janzen (1976) cited records of Burmese jungle fowl and African monkeys feeding on bamboo flowers. Insects may damage bamboo florets ()ohn <7 al. 1995, Koshy & Harikumar 2001). This suggests yet another dimension to the vulnerability' of long- lived bamboos during the brief phase of sexual reproduction that precedes death. The value of highly synchronised flowering in satiating consumers, as postulated by the seed predator satiation hypothesis , may also apply to 'predators of flowers'. Acknowledgements My study of Bambusa amhemica was funded by the Parks & Wildlife Commission of the Northern Territory. At the Territory Wildlife Park, Jasmin Jan gave permission to conduct the trials, and Greg Mayo facilitated them. Peter Whitehead commented helpfully on a draft. References Bhargava A., Kumbhare V., Srivastava A. and Sahai A. (1996) Bamboo parts and seeds for additional source of nutrition .Journal of Food Science and Technology (Mysore) 33, 145-146. Chauhan N.S. and Saxena R.N. (1985) The phenomenon of bamboo flowering and associated increase in rodent population in Mizoram. Journal of the Bombay Natural History Society 82, 644-647. Curran L.M. and Webb C.O. (2000) Experimental tests of the spatiotemporal scale of seed predation in mast-fruiting Dipterocarpaceae. EcologicalMonograpbslO, 129-148. Dostine P.L. and Franklin D.C. (2002) A comparison of the diet of three finch species in the Yinberrie Mills area. Northern Territory. Emu 102,159-164. Franklin D.C. (2003) Morphology and taxonomy of the Top End Bamboo Bambusa amhemica F. Muell., a little-known bamboo from northern Australia. Bamboo Science and Culture 17, 44-54. Franklin D.C. (2004) Synchrony and asynchrony: observations and hypotheses for the flowering wave in a long-lived semelparous bamboo. Journal of Biogeography 31, 773-786. Higgins P.J. (1999) Handbook of Australian, New Zealand and Antarctic Birds. Ho/nme 4. Parrots to Dollarbird. Oxford University Press, Melbourne. Immelmann K, (1982) Australian Finches. Angus & Robertson, Sydney. Jackson H.D. (1972) The status of the pied mannikin, Ijtnchura fringilloides (La fresnaye) in Rhodesia and its association with the bamboo Oxytenanthera abyssinica. Rhodesian Science News 6,342-343. Jaksic F.M. and Lima M. (2003) Myths and facts on ratadas: Bamboo blooms, rainfall peaks and rodent outbreaks in South America. Austral Ecology 28,237-251. Janzen D.H. (1976) Why bamboos wait so long to flower. Annual Review of Ecology and Systematics 7, 347-391. John C.K., Nadgauda R.S. and Mascarenhas A.F. (1995) Floral biology and breeding behaviour in Bambusa amndinacea. Journal of Cytology and Genetics 30, 101-107. Kelly D. (1994) The evolutionary ecology' of mast seeding. Trends in Ecology and Evolution 9, 465- 470. Koshy K.C. and Harikumar D. (2001) Reproductive biology' of Ocblandra scriptoria , an endemic reed bamboo of the Western Ghats, India. Bamboo Science and Culture 15, 1-7. Pathak N.N. (1979) Nutritive value of bamboo ( Bambusa sp.) grains for mice. Indian Journal of Nutrition and Dietetics 16, 356-358. 50 Northern Territory Naturalist 18 (2005) D. Franklin Read J.L. (1994) The diet of three species of firetail finches in temperate South Australia. Em/t 94, 1-8. Sakai S. (2002) General flowering in lowland mixed dipterocarp forests of South-east Asia. Biologicaljournal of the I jnnean Society 75, 233-247. Todd M.K., Felton A. and Garnett S.T. (2003) Morphological and dietary differences between common and uncommon subspecies of Crimson Finch, Neoc/imia phaeton , and Star Finch, Neochm'ta mftcauda , in northern Australia. Emu 103, 141-148. Seed-eating birds avidly consume flowers of the Top End Bamboo Bambusa amhemica. (Don Franklin) Northern Territory Naturalist (2005) 18: 51-53 A population count and ecological notes for the little-known terrestrial orchid Didymoplexis pallens Donald C. Franklin 1 and Raelee Kerrigan 2 1 School for Environmental Research, Charles Darwin University, Darwin, NT 0909. Email: don.franklin@cdu.edu.au 2 Northern Territory Herbarium, Department of Infrastructure, Planning and Environment, PO Box 496, Palmerston, NT 0831. Email: raelee.kerrigan@nt.gov.au Didymoplexis pollens Griff, is a leafless, saprophytic, terrestrial orchid. Its small white flowers open one at a time in a raceme atop an unbranched, 10—15 cm, pink, fleshy stem. For most of the year it persists underground as a tuber, emerging to flower and fruit following early wet season rains (|ones 1988). The species is known from Queensland, Northern Territory, Western Australia, Indonesia, Malaysia and India, but is considered "elusive" (Hooker 1894, Jones 1988). In the Northern Territory (NT), Liddle et ol, (1994) mapped its occurrence at two locations in the Daly River catchment and two in central Arnhem Land. It has since been located on Melville Island (NT Herbarium record). On the basis of the sparsity of records and limited and anecdotal information about its occurrence, the species has been classified for the NT as Data Deficient (DIPE 2002). In this note, we provide details of a sixth NT location for D. pollens including a population count and brief ecological notes. The population was discovered by one of us (Franklin 2003) during the course of a study of the Top End Bamboo Bombnso omhemica at Mar)' River Park (12°54'30" S, 131°39'30" E), a private ecotourism facility, in November 2001. At the time, an estimated 100 individuals were found growing under bamboo clumps on the bank of a billabong embedded in the floodplain of the Maty River approximately one kilometre upstream from the Arnhem Highway bridge over the Mary River. On 2 December 2003, we conducted a search for the species over about 3 km of billabong bank at Maty River Park, including the area of the original discovery. We located D. pollens intermittently over a 1.2 km length of bank, and counted exactly 700 individuals, of which an estimated 90% were in flower and 10% in fruit. Most D. pollens were found growing in bamboo leaf litter. Some were under other riparian trees, shrubs or vines — Phyllanthus reticidatus , Fiats scobina , \jophostemon lactijluus and Gymnonthero oblonga, and a few under a light cover of the grass Faspalidinm distans that occurs in open areas immediately above the riparian forest. The soils supporting D. pallens were alluvial, grey clay-loams and, perhaps surprisingly for soils supporting a saprophyte, not notably humic. All sites are prone to inundation (Franklin, pers. 52 Northern Territory Naturalist 18 (2005) D. Franklin & R. Kerrigan obs.), most often during the mid- to late-wet season after flowering and fruiting have finished. We did not locate any D. pallens in the lower slope of the riparian forest where inundation may be prolonged and occurs earlier in the season. Didymoplex/s pallens tubers and presumably also seed are thus tolerant of waterlogging, but excess and/or early waterlogging may limit its occurrence, a strikingly similar ecological pattern to that of the bamboo under which it occurs at Mary River (Franklin & Bowman 2004). Didymopkxis pallens is not exclusively associated with bamboo. Notes accompanying previous herbarium collections for the NT indicate occurrence in a variety of damp and/or poorly-drained situations, including a seepage area on the margin of monsoon rainforest in sandstone country', amongst sedges at the edge of a spring-fed monsoon rainforest, in leaf litter in vine forest along a river, and on a swampy plain treeless but for some Pandanus. Reasons for the patchiness and scarcity of D. pallens in the NT, its response to fire, and the threats facing it, if any, are speculative. The Mary River site was subject to an intense wildfire in August 2001, and D. pallens was observed on burnt ground in the following November (Franklin 2003). The area was not burnt for at least several years prior to 2001, and has not been burnt since. Tuberous plants are particularly susceptible to disturbance by Feral Pigs Sus scrofa (e.g. Fensham 1993) and possibly also Asian Water Buffalo Bnbalus bubalis. Herbarium notes associated with one of the Arnhem Land sites report heavy disturbance by buffalo. Under current management, there are no buffalo in the riparian forest areas of Maty River Park. An active control program for feral pigs has been successful in reducing them to low levels in the area where we found D. pallens , as evidenced by notably low levels of impact on bamboo shoots compared to other sites (Franklin, unpubl.). Based on the available evidence of limited extent of occurrence and apparently small population, a case could be made for classifying D. pallens as Vulnerable in the NT. However, plants are evident above ground for just a few weeks each year (Franklin 2003) and at a time when the weather is particularly unfavourable for field surveys, so it may readily be overlooked in other areas. Annua! fluctuations in the number of orchids that emerge seasonally from tubers (Gillman & Dodd 1998) may overstate real population fluctuations because of tuber dormancy (Brzosko 2003, Kery & Gregg 2004) . We therefore recommend that D. pallens continue to be classified as Data Deficient in the NT, and that further survey and study of the population dynamics of, and threats to the species, be undertaken. Acknowledgements We are grateful to the proprietors of Mary' River Park - Mike and Gina Ostwald and Emma Kendall - for their permission to survey the population. Glenn Wightman, Sheryl Keates and Jarrad Holmes assisted with the count. Ian Cowie and an anonymous referee commented constructively on a draft of the manuscript. Orchid count Northern Territory Naturalist 18 (2005) 53 References Brzosko E. (2003) The dynamics of island populations of Platanthera bifolia in the Biebrza National Park (NE Poland). Anttaks Botanici Fenrnci 40, 243-253. Department of Infrastructure Planning and Environment. (2002) Threatened species of the Northern Territory. Information Package. 2002. Northern Territory Government, Darwin. Fensham RJ. (1993) The impact of pig rooting on populations of Bunr/annia sp., a rare rainforest herb on Bathurst Island. Proceedings of the Royal Society of Queensland 103, 5-12. Franklin D. (2003) Rare orchid at Mary River Park. Nature Territory] une 2003, 3. Franklin D.C. and Bowman D.M.J.S. (2004) A multi-scale biogeographic analysis of Bambusa amhemica , a bamboo from monsoonal northern Australia. Journal of Biogeography 31, 1335- 1353. Gillman M.P. and Dodd M.E. (1998) The variability of orchid population size. Botanical Journal of the Unnean Society 126, 65-74. Hooker J.D. (1894) The Flora of British India. 1 'ohmic 6. Orcbidaceae to Cyperaceae. L. Reeve, Ashford, Kent, U.K. Jones D.L. (1988) Native Orchids of Australia. Reed, Sydney. Kery M. and Gregg K.B. (2004) Demographic analysis of dormancy and survival in the terrestrial orchid Cypripeditim reginae. Journal of Ecology 92, 686-695. Liddle D.T., Russell-Smith J., Brock J., Leach G.J. and Connors G.T. (1994) Atlas of the Vascular Rainforest Plants of the Northern Territory. Australian Biological Resources Study, Canberra. Northern Territory Naturalist (2005) 18: 54-60 Birds of Lake Nongra and surrounding bushiand, Northern Territory Martin Gole 8 Landor Road, Gooseberry Hill, WA 6076. Email: mgole@westnet.net.au Lake Nongra is a large freshwater lake situated in the southern-most part of the pastoral country on the northern margin of the Tanami Desert in the Northern Territory. The lake is located at the northern edge of range for many arid zone birds and at the southern edge of range for many sub-tropical zone birds. The area is rarely visited and it appears that the last comprehensive bird survey of the lake was undertaken in 1993, when 29 waterbird species were recorded (Jaensch 1994). The birds reported here were recorded during eight surveys conducted over a 23-day period during October-November 2002. In addition, a brief survey was undertaken in June 2002 when water levels were about 1 m higher than in October-November. I observed 17 waterbirds not previously recorded, did not record 5 seen previously and also recorded 54 bushbirds around the lake foreshore. The lake, when full, is shallow with a maximum depth of 1.5—2 m. It was last dry about 10 years ago and is currently only 0.3 m deep and very restricted in areal extent (T. Westerway, Inverway Station, pers. comm.. Sept. 2005). It is filled by summer rains. During the October-November surveys the lake was less than at full capacity (Figure 1), being at its widest points 9 km north-south and 5—6 km east-west and was dominated by open water. In places patches of thick, dead bushes, 2-4 m high, extended 100—200 m into the lake. These occur particularly in the north-west but also along parts of the north-eastern and eastern shoreline and along creek inlets into the lake. The lake margin, 50-500 m wide, is flat and during the survey period was covered by short, ephemeral grasses and herbs except for 2—30 m wide mudflats exposed along the edge of the lake and along some creek channels. The surrounding bush is dominated by open eucalvpt and acacia woodland with an understory of grass although along creek lines die vegetation is more complex and relatively dense. The lake straddles the boundary' of Inverway and Birrindudu pastoral leases. On Inverway the lake is fenced off from cattle, however on Birrindudu the paddock containing the lake is used as a drought refuge and is usually not grazed except by stray cattle. Few recent cattle tracks were seen along the shoreline. However, due to dry conditions there were plans to re-introduce cattle into this paddock. Several camels were seen drinking from the lake during the October-November surveys. The eight surveys were each of 1.5—2.5 hour duration and covered an area of about 500 m radius with the centre point taken on the edge of the lake except for the southern-most site which was along a creek line (Figure 1). Large parts of the lake Lake Nongra birds Northern Territory Naturalist 18 (2005) 55 were not surveyed. In addition, the thickets of dead bushes present in places around the lake prevented the observation of probably many hundreds of waterbirds sheltering from windy conditions that prevailed during many of the surveys. Leica xlO binoculars were used for observations, so for the smaller waterbirds only those relatively close to the shore could be identified. In June most of the many hundreds of waterbirds seen at the single site visited were well off-shore and were not identified. Because of its incomplete nature only one species from this survey (Blue-billed Duck), not seen in the later surveys, is included in Table 1. During the October-November surveys most waterbirds were sheltering close to the shore from the windy conditions and a high proportion of birds observed within each survey area were identified. Figure 1. Location of survey sites around the margin of Lake Nongra undertaken in October-November 2002. The heavy 7 oudine of the lake is its approximate position in October 2002 whereas the dotted line is the lake outline as shown on the 1:100 000 scale topographic map. 56 Northern Territory Naturalist 18 (2005) M Gole Table 1. Birds seen at Lake Nongra between 14/10/02 and 5/11/02. Bird Species Number of surveys (n = 8) Magpie Goose Anseranas semipalmata 1 Blue-billed Duck Oxyura australis 1 Freckled Duck Stictonetta naevosa 3 Black Swan Cygnus atratus 3 Australian Wood Duck Chenonetta jubata 1 Pacific Black Duck Anas superciliosa 7 Grey Teal Anas gracilis 8 Pink-eared Duck Malacorhynchus membranaceus 7 Hardhead Aythya australis 8 Australasian Grebe Tachybaptus novaehollandiae 1 Great Crested Grebe Podiceps cristatus 3 Darter Anhinga melanogaster 7 Little Pied Cormorant Phalacrocorax melanoleucos 3 Little Black Cormorant Phalacrocorax sulcirostris 6 Australian Pelican Pelecanus conspicillatus 8 White-faced Heron Egretta novaehollandiae 3 Little Egret Egretta garzetta 4 White-necked Heron Ardea pacifica 2 Great Egret Ardea alba 7 Glossy Ibis Plegadis falcinellus 8 Straw-necked Ibis Threskiomis spinicollis 3 Royal Spoonbill Platalea regia 5 Yellow-billed Spoonbill Platalea ftavipes 5 Black-shouldered Kite Elanus axillaris 1 Whistling Kite Haliastur sphenurus 5 Spotted Harrier Circus assimilis 1 Swamp Harrier Circus approximans 1 Brown Goshawk Accipiter fasciatus 1 Collared Sparrowhawk Accipiter cirrhocephalus 1 Wedge-tailed Eagle Aquila audax 2 Brown Falcon Falco berigora 2 Australian Hobby Falco longipennis 2 Nankeen Kestrel Falco cenchroides 2 Brolga Crus rubicunda 4 Purple Swamphen Porphyrio porphyrio 1 Black-tailed Native-hen Gallinula ventralis 4 Eurasian Coot Fulica atra 4 Australian Bustard Ardeotis australis 2 Marsh Sandpiper Tringa stagnatilis 3 Common Greenshank Tnnga nebularia 5 Wood Sandpiper Tnnga glareola 1 Common Sandpiper Actitis hypoleucos 2 Black-winged Stilt Himantopus himantopus 8 Red-capped Plover Charadnus ruficapillus 2 Oriental Plover Charadnus veredus 1 Black-fronted Dotterel Elseyornis metanops 4 Red-kneed Dotterel Erythrogonys cinctus 7 Lake Nongra birds Northern Territory Naturalist 18 (2005) 57 Bird Species Number of surveys (n = 8) Masked Lapwing Vanellus miles 5 Gull-billed Tern Sterna nilotica 7 Caspian Tern Sterna caspia 2 Whiskered Tem Chlidonias hybridus 7 Crested Pigeon Ocyphaps lophotes 4 Peaceful Dove Ceopelia striata 2 Red-tailed Black-Cockatoo Caiyptorhynchus banksii 4 Galah Cacatua roseicapilla 1 Little Corella Cacatua sanguinea 3 Australian Ringneck Bamardius zonarius 2 Budgerigar Melopsittacus undulatus 1 Pallid Cuckoo Cuculus pallidus 1 Pheasant Coucal Centropus phasianinus 1 Tawny Frogmouth Podargus stn'goides 1 Sacred Kingfisher Todlramphus sanctus 2 Rainbow Bee-eater Merops omatus 2 Red-backed Fairy-wren Malurus melanocephalus 4 Striated Pardalote Pardaiotus striatus 1 Weebill Smicrornis brevirostris 4 Spiny-cheeked Honeyeater Acanthagenys rufogularis 1 Little Friarbird Philemon citreogularis 4 Yellow-throated Miner Manorina flavigula 5 Singing Honeyeater Lichenostomus virescens 4 Grey-fronted Honeyeater Lichenostomus plumulus 4 Yellow-tinted Honeyeater Lichenostomus flavescens 2 Brown Honeyeater Lichmera indistincta 4 Rufous-throated Honeyeater Conopophila rufogularis 4 Jacky Winter Microeca fascinans 4 Grey-crowned Babbler Pomatostomus temporalis 2 Rufous Whistler Pachycephala rufiventris 5 Grey Shrike-thrush Colluricincla harmonica 1 Restless Flycatcher Myiagra inquieta 5 Magpie-lark Grallina cyanoleuca 8 Willie Wagtail Rhipidura leucophrys 7 Black-faced Cuckoo-shrike Coracina novaehollandiae 6 White-winged Triller Lalage sueurii 7 White-breasted Woodswallow Artamus leucorynchus 8 Black-faced Woodswallow Artamus cinereus 8 Little Woodswallow Artamus minor 1 Pied Butcherbird Cracticus nigrogularis 7 Australian Magpie Gymnorhina tibicen 1 Torresian Crow Con/us orru 3 Singing Bushlark Mirafra javanica 1 Richard's Pipit Anthus novaeseelandiae 4 Zebra Finch Taeniopygia guttata 3 Tree Martin Hirundo nigricans 1 Fairy Martin Hirundo ariel 7 Rufous Songlark Cincloramphus mathewsi 1 58 Northern Territory Naturalist 18 (2005) M Gole All 41 waterbirds and 54 bushbird species observed are listed in Table 1. Taxonomic order and nomenclature follow Christidis and Boles (1994). Waterbirds The most common species seen were Grey Teal and Hardhead with estimated numbers of >100 to >1000 in most surveyed areas. Pacific Black Duck was also common with numbers of 50 to >500 per site. Black Swan, Pink-eared Duck, Darter, Australian Pelican, Royal Spoonbill and Yellow-billed Spoonbill all occurred in similar numbers of 10-20 at most sites. Moderate numbers (75 to >150) of Eurasian Coot occurred in four of the northern sites. Freckled Duck were seen only in the three most nordterly sites in low' numbers (10-20). Freckled Duck are regularly reported from Lake Gregory', 330 km southwest of Lake Nongra and situated in a similar arid to sub-tropical transition zone (Halse et al 1998). They also occur at Lake Woods, 400 km to the east. They are, how'ever, rarely reported from elsewhere in the northern part of Australia (Marchant & Higgins 1990, Barrett et al. 2003). Red-kneed Dotterel w r as the most common w'ader and along with Black-w'inged Stilt occurred in numbers of 10 to >30 in all except one site. Other waterbirds present at most survey sites were Glossy Ibis (2-3 birds per site), Great Egret (0—5), Little Black Cormorant (0—10), Whiskered Tern (0—15), Gull-billed Tern (0—3), Common Greenshank (0—5), and Black-tailed Native-hen (10-30). All other waterbirds w'ere present as single birds or small numbers at one or two sites (Table 1). Of particular interest is the presence of male and female Blue-billed Ducks on 27 June 2002 in the north-east of the lake with the male in breeding plumage. Blue-billed Duck is rarely recorded in the arid interior and the Top End of Australia. None have been reported from Lake Gregory (Halse et al. 1998). Marchant and Higgins (1990) report only one Northern Territory record and the nearest records in Barrett et al. (2003) are at Alice Springs and Glen Helen Gorge in central Australia. The record for Magpie Goose extends its know'n southern distribution in this part of the Northern Territory although they have been recorded from just across the border in Western Australia (Barrett et a/. 2003) and also occur at Lake Gregory (Halse et al. 1998). The presence of Great Crested Grebe and Purple Swamphen at Lake Nongra adds an additional locality to their sparse distribution in the Northern Territory and arid northern Western Australia (Marchant & Higgins 1990, 1993, Barrett et al 2003). The other waterbirds recorded have widespread distributions and their presence at Lake Nongra is not unexpected. The only birds observed breeding w'ere a pair of Masked Lapwing (2 eggs). Waterbirds not recorded by me but reported by Jaensch (1994) are Plumed Whistling- Duck, Nankeen Night Heron, Australian White Ibis, Baillon’s Crake, and Clamorous Reed-Warbler. The latter tw'o species were seen in the north-west part of the lake, an area not visited by me. Lake Nongra birds Northern Territory Naturalist 18 (2005) 59 The total number of waterbirds recorded by Jaensch (1994) was 1880. It is not possible from my observations to estimate waterbird numbers for the whole lake, however it is clear that there were many more birds present than during the 1993 survey. During the first of the southern October-November 2002 surveys there were about 2000 waterbirds in the southernmost ~1.0 km of the lake. Many additional birds are likely to have been present elsewhere in the lake, particularly in the north¬ west part where several major creeks enter and where trees appear to extend well into the lake. The apparendy low numbers of waterbirds recorded by Jaensch (1994) at Lake Nongra in early-mid 1993 corresponds to the lowest numbers recorded at Lake Gregor)' in the period 1988 to 1995 (4376 in 1993 compared to 44 141-650 000 in seven other surveys, Halse et al. 1998). This was a wet year and birds were likely to have been widely dispersed. This suggests that higher numbers than recorded by Jaensch (1994) might normally be expected at Lake Nongra, at least when the lake is not dry. Bushbirds Most bushbirds recorded around the lake margins have widespread distributions. However, Red-tailed Black-Cockatoo (subspecies macrorbyncbus), Little Friarbird, Yellow-tinted Honeyeater and Rufous-throated Honeyeater are at or close to the known inland edge of their range. The lake also marks the approximate southern boundary for Grey-fronted Honeyeater, Jacky Winter and Restless Flycatcher which do not extend into the Tanami Desert, although these birds also occupy more southerly regions elsewhere. Lake Nongra is close to the known northern limit for Spiny-cheeked Honeyeater at this longitude (Blakers et al. 1984, Barrett et al. 2003). Lake Nongra is a large isolated wedand within a semi-arid terrain and probably has regional significance for waterbirds. The apparently high numbers of waterbirds present in 2002 compared to 1993 may reflect the use of the lake as a drought refuge given the very dr)' conditions experienced in the area during 2001-2002. Seventy- three waterbirds have been recorded from Lake Gregory, a similarly situated and regionally significant wedand, so future surveys at Lake Nongra are likely to record many additional species. References Barrett G., Silcocks A., Barry S., Cunningham R. and Poulter R. (2003) 'The New Atlas of Australian Birds. Birds Australia, Melbourne. Blakers M., Davies S.J.J.F. and Reilly P.N. (1984) The Atlas of Australian Birds. RAOU and Melbourne University Press, Melbourne. Christidis L. and Boles W.E. (1994) ’I'he Taxonomy and Species of Birds of Australia and its Territories. Royal Australian Ornithologists Union Monograph 2, RAOU, Melbourne. 60 Northern Territory Naturalist 18 (2005) M Gole Halse S.A., Pearson G.B. and Kay W.R. (1998) Arid zone networks in time and space: waterbird use of Lake Gregory in north-western Australia. International Journal of Ecology and Environmental Sciences 24, 207-222. Jaensch R.P. (1994) An inventory' of wetlands in the sub-humid tropics of the Northern Territory'. Report to the Australian Nature Conservation Agency, Conservation Commission of the Northern Territory, Darwin. Marchant S. and Higgins P.J. (eds) (1990) Handbook of Australian, New Zealand and Antarctic Birds, Volume 1, Ratites to Ducks. Oxford University Press, Melbourne. Marchant S. and Higgins P.J. (eds) (1993) Handbook of Australian, New Zealand and Antarctic Birds, Volume 2, Raptors to lutpnnngs. Oxford University Press, Melbourne. Although common in coastal areas of the Northern Territory, the Magpie Goose Anseranas semipalmata is infrequently seen as far inland as Lake Nongra. (Martin Armstrong) Northern Territory Naturalist (2005) 18: 61-67 Use of tree hollows by the Green Tree Frog Litoria caerulea at East Point Reserve, Darwin S.J. Reynolds School of Science and Primary Industries, Charles Darwin University, Darwin, NT 0909. The Green Tree Frog IJtoria caerulea has one of the broadest geographical ranges of all Australian hylids. Its distribution extends in a wide arc from south of Broome (Anna Plains) through the Kimberley, across the Top End south and inland to the Tanami, Sturt Plateau and Barkly Tableland, and down the east coast through Queensland to northern New South Wales (Barker el al. 1995, Tyler el al. 1983, P. Horner, pers. comm.). It is also extralimital in New Guinea (Tyler 1999). The three other species in the IJtoria caerulea complex each have relatively restricted distributions. IJtoria gilleni occurs in the central Australian ranges (MacDonnells), IJtoria splendida in the East Kimberley and Keep River region, and IJtoria cavemicola is a habitat specialist confined to the Mitchell Plateau of the Kimberley. IJtoria gilleni was originally described by Spencer (1896) and is currently considered a good species (e.g. Cogger et al. 1983, Bedford 2000), as are K cavemicola (Tyler & Davies 1979) and /. splendida (Tyler et al 1977). Tyler and Davies (1986) included I - gilleni under K caerulea which was noted as occurring "throughout the Northern Territory". Whilst the three geographically restricted members of the IJtoria caerulea complex are largely or exclusively rock dwelling (saxicoline) species that use moist microclimates in rock crevices (often near sources of permanent water) as retreats, K caendea uses tree hollows as shelter sites across its range. It occurs in a vide range of habitats including savanna, pindan, eucalypt woodlands and forests, closed monsoon forests and thickets, and has also been recorded from mangroves (J. Smith, pers. comm.) and mango orchards (R. Peng, pers. comm.). In many parts of Australia, including the Top End, it has become accustomed to human habitation and commonly breeds in garden ponds. As part of ongoing physiological studies of Top End amphibians, I have been involved in investigations of the biology’ of I - caemlea, with the focus of the work at East Point Reserve near Darwin. The Reserve includes a mosaic of grassed and revegetated areas, with a single patch of relatively undisturbed coastal rainforest occupying an area of approximately 25 ha within which the studies have been conducted. The vegetation at the study site is defined as a low closed coastal dry’ monsoon forest {sensu Russell-Smith 1991). It consists of a mixed array of trees, shrubs and vines including many Malesian floral elements and generally extends to a height of 9—12 metres. Conditions under the dense rainforest canopy are highly’ suited 62 Northern Territory Naturalist 18 (2005) S. Reynolds to tree frogs, incorporating extensive shade, warm and relatively constant temperatures, high humidity, and minimal air movement particularly near the forest floor (McCay 2003). These last two factors in particular serve to reduce rates of evaporative water loss (e.g. Christian & Parry' 1997, Tracy 1976) in what is a seasonally dry' (monsoonal) and therefore potentially highly desiccating environment. At East Point the frogs are primarily active during the wet season. They are probably entirely nocturnal, emerging soon after dusk and moving to the canopy, the forest floor or low perches on branches. Breeding takes place opportunistically during the wet season, calling activity being particularly' intense during heavy rain. Frogs tend to remain within tree hollows during the coolest months of the dry season (G. Miles, pers. comm.), but are relatively active when weather conditions are suitable, i.e. during periods of high night time humidity' or warmer weather. One aspect of the investigations at East Point has been to examine the use of daytime refuges by L. caemlea. Individual frogs have been followed to their daytime refuges with the aid of radio-tracking (telemetry) techniques, and a pattern of use of hollows has emerged. Frogs that were tracked were given a unique identification number or code, and tree hollows were numbered after a frog had been followed and the location of the hollow determined. In certain instances, where hollows were high above the ground, the tree species could be identified but it was not possible to measure the attributes of the hollow. Tree species that form hollows were identified using Brock (1997), Wightman and Andrews (1989) and Booth et al (2001) and with reference to collections maintained at the Northern Territory' (NT) Plerbarium. A list of plant species for the area was derived from NT Herbarium records and Wightman and Andrews (1989). Based on this information, approximately 140 plant species occur in the monsoon forest at East Point (excluding artificial plantings). This includes 48 species of tree (defined as plants with a robust trunk and capable of exceeding five metres in height), only nine of which are utilised by the frogs as daytime refuges (Table 1). Several trees of lJtsea glutinosa, Drypetes deplanchei and Ganophyllumfalcatum were used as hollows, whereas for the remaining six species there were only single cases of hollow use (Table 1). Only certain types of trees in the monsoon forest readily form hollows, and of these only some appear to be suitable as refuge sites for frogs. 1 follows vary in structure as a result of their mode of formation. In Utsea glutinosa knots form at regular intervals along the trunk at previous points of attachment for branches and become small hollows; hollows also form in the trunk of dead individuals of this species (Table 1). Convolutions of the trunk in Drypetes deplanchei form a vertical envelope or fissure of varying extent which may become almost entirely enclosed. Hollows of this type tend to be larger but probably do not provide as much protection as closed hollows. Ganophyllum falcatum forms cylindrical hollows along main branches, some of which are many metres above the ground. Dead and decaying trees are also used and termites Tree Frog hollows Northern Territory Naturalist 18 (2005) 63 are an important agent of hollow formation. However, our data indicate that trees are not utilised bv frogs after they have fallen to the ground. Table 1. Characteristics of trees with hollows that were utilised as daytime refuges by \Jtoria caemlea at East Point Reserve (« — 25 hollows). (H = Hollow number, Frog ID = identification number or code of frog/s that used the hollow) H Frog ID Tree species DBH (cm) Height (m) 1 10, X Litsea glutinosa 10 9 7 D Litsea glutinosa 6 7 22 11 Litsea glutinosa 10 6 25 11 Litsea glutinosa 25 10 26 11 Litsea glutinosa 15 9 27 11 Litsea glutinosa 12 12 2 A Drypetes deplanchei 22 10 3 B Drypetes deplanchei 20 10 5 C Drypetes deplanchei 25 11 9 9 Drypetes deplanchei 20 9 23 11 Drypetes deplanchei 21 9 6 10, 8 Ganophyllum falcatum 60 14 10 2 Ganophyllum falcatum 30 11 11 8 Ganophyllum falcatum 50 10 16 8 Ganophyllum falcatum 30 12 18 11 Denhamia obscura 16 9 20 15, 13 Pouteria sericea 12 9 14 8 Miliusa brahei 20 10 17 11, X Polyalthia nitidissima 16 9 15 7, 16 Strychnos lucida 20 9 21 7 Acacia auriculiformis 40 16 4 2, 12 dead Litsea 7 7 8 4 mostly dead 20 11 19 17 termite ridden 8 7 X X dead slender tree (Litsea ?) 10 5 64 Northern Territory Naturalist 18 (2005) S. Reynolds Trees used by Green Tree Frogs display a wide range of characterisdcs, from tall trees with dense foliage to low spreading species (Table 2), aldrough the majority' of trees are greater than seven metres in height (Table 1). There are also a range of bark types from smooth to fissured. As in other dry monsoon forests of the Northern Territory, a subset of the vegetation is deciduous during the dry' season (Bach 2002). As a consequence, vegetative cover is less adequate and there is greater penetration of sunlight and potential for air movement during the dry season. These factors are likely to increase the importance of tree hollows as daytime refugia during this period. Table 2. Characteristics of hollow-forming tree species used by IJforia caerulea. Tree Species Max. height (m) Habit Bark Foliage Leaf retention Acacia auriculiformis 20 tall spreading rough; fissured at base dense evergreen Denhamia obscura 10 rounded crown rough pendulous evergreen Drypetes deplanchei 12 upright buttressed smooth to slightly rough dense semi- deciduous Ganophyllum falcatum 20 large tree smooth to slightly coarse; flaking dense evergreen Litsea glutinosa 15 slender upright smooth to slightly rough moderately dense deciduous ?facultatively Miliusa brahei 15 erect rough, fissured moderately dense deciduous Polyalthia nitidissima 20 upright slightly rough dense evergreen Pouteria sericea 10 erect rough, finely fissured moderately dense evergreen Strychnos lucida 6 low spreading smooth to slightly rough moderately sparse facultatively deciduous Hollows used by /, caerulea vary considerably in depth, from 8 to 30 cm, but the width of the opening is consistently less than 4 cm (Table 3). The majority of trees used are greater than 9 cm diameter at breast height (DBH: Table 1). Hollows are generally at head height or above, although they may be quite low, eg. Hollow 1 (Table 3). I have Tree Frog hollows Northern Territory Naturalist 18 (2005) 65 only managed to record measurements for a subset of the hollows because several were inaccessible or the exact location could not be identified using radio-tracking techniques (Table 3), hence it is likely that the data are biased in favour of hollows nearer the ground (cf. Griffiths 1994). Consideration of the microhabitat requirements of I _ caemlea suggests that optimal hollows are likely to be those that retard moisture loss and secondarily may also reduce vulnerability to predators and competitors. A tight-fitting hollow is presumably optimal for maintaining moisture balance and in some situations the head of the animal may block the hollow; this may serve to reduce water loss and may also discourage predators. Potential predators at East Point include Children’s Python IJasis children Common Tree Snake Dendrelapbis pnnctulata, Slaty-grey Snake Stegonotus cumllatus, the monitor lizards Varanus panopfes and V. scalaris, and Pacific Baza Aviceda suberistata. Some hollows retain water during the wet season and the frogs have been observed to conceal themselves underwater when approached. Frogs also use man¬ made structures (ablution blocks) at East Point, which provide suitable shelter and moisture conditions. Table 3. Characteristics of tree hollows used by Utoria caerulea at East Point Reserve. Tree species are listed in Table 1. Hollows 20u, 16, 21, and additional hollows (not listed) were inaccessible or could not be located precisely. 1 and u denote lower and upper respectively. ollow No. Height above ground (m) Width of opening (cm) Hollow depth (cm) Orientation Hollow Type 11 6 4 10 diagonal branch 151 4 3 10 horizontal trunk 15u 4.5 2.5 8 vertical trunk 4 3 4 9 vertical dead hollow trunk 1 0.6 1.5 4 vertical slit/knot 17 1.6 2 10 vertical in fork 2 2 1.5 30 vertical fissure 23 1.5 2 8? vertical fissure 201 4 2 8 vertical trunk knot 20u 4+ ? ? ? knot ? 26 2.1 2 8 vertical in fork 16 5+ ? ? ? branch ? 21 5+ ? ? ? ? X 1.1 5 20+ vertical hollow trunk 66 Northern Territory Naturalist 18 (2005) S. Reynolds On two occasions we have recorded two individuals using different hollows in the same tree, and at times we have observed two frogs using the same hollow. The choice of hollow may indicate individual preference for particular tree species; certainly it appears to be the case that distinct hollow types are selected by individual animals. A visual search of potentially suitable hollows was made in a portion of the forest, and a low level of occupancy was found, suggesting that frogs are selecting particular types of hollows as shelter sites. Surveys of tree frequency (unpubl. data) indicate that of the commonly utilised species lJtsea glutinosa is relatively abundant in the forest patch at East Point, whereas, for example, GanopbyUum fakatum occurs at low densities. Also, although stem densities are exceedingly high (approximately 9000/ha), average DBH is low (Mean ± SD (cm): 5.55 ± 3.98) with few stems > 9 cm DBH (~20%), and of these only some appear to have the potential to fonn hollows. Green Tree Frogs appear also to use tree hollows in woodland habitats (often near water), but there is no data on the types of shelter sites in the range of other habitats that they occupy. In addition, I am yet to confirm whether (as I suspect) densities are highest in areas of moist microclimate such as rainforest patches. Finally, although Green Tree Frogs are widespread, common and frequently encountered, this study represents what is effectively the limit of the knowledge of the ecology of the species. Further research is required to describe the basic biology of this species, and indeed much of the Torresian herpetofauna. Acknowle dgements Keith Christian supported the work and critiqued an earlier draft. Lorrae McArthur laid the foundation for work on GTFs at East Point. Chris Tracy and Matthew Shortus were involved in the fieldwork. Bob Harwood (NT Herbarium) furnished plant identifications and Andrea Hope provided a list of plant species at the Reserve. Angelika Hesse (Darwin City Council) provided information on East Point Reserve. References Bach C.S. (2002) Phenological patterns in monsoon rainforests in the Northern Territory, Australia. Austral Ecology 27, 477-489. Barker J., Grigg G.C. and Tyler M.J. (1995) A Field Guide to Australian Froqs. Surrey Beatty' & Sons, NSW. Bedford G. (2000) Colour change during mating in the arid zone frog I Jtoria gi/leni. Herpetofauna 30, 51. Booth R., Harwood R.K. and Mangion C.P. (2001) Field Key for the Monsoon Rainforest Flora of the Darwin Region. Northern Territory' University Press, Darwin. Brock J. (1997) Natire Plants of Northern Australia. Reed Books, Victoria. Christian K. and Parry' D. (1997) Reduced rates of water loss and chemical properties of skin secretions of the frogs / Jtoria caerulea and Cydorana australis. Australian Journal of Zoology 45, 13-20. Tree Frog hollows Northern Territory Naturalist 18 (2005) 67 Cogger I I.G., Cameron li.li. and Cogger H.M. (1983) Zoological Catalogue of Australia Volume 1 Amphibia and Reptilia. Australian Government Publishing Sendee, Canberra. Griffiths A.D. (1994) The effect of seasonal environment and fire on the ecology of frillneck lizards, Ch/amydosaums kingii , in the wet-dry tropics of northern Australia. M.Sc. Thesis, Northern Territory University'. McCay M.G. (2003) VC inds under the rain forest canopy: the aerodynamic environment of gliding tree frogs. Biotropica 35, 94-102. Russell-Smith J. (1991) Classification, species richness, and environmental relations of monsoon rain forest in northern Australia. Journal of Vegftation Science 2, 259-278. Spencer B. (1896) Report on the work of tlx Horn Scientific Expedition to central Australia. Pt. 2 Zoology: Amphibia. London. Tracy' C.R. (1976) A model of the dynamic exchanges of water and energy between a terrestrial amphibian and its environment. Ecological Monographs 46, 293-326. Tyler M.J. (1999) Distribution Patterns of Amphibians in the Australo-Papuan Region. In Patterns of Distribution of Amphibians (ed. W.E. Duellman), pp. 541-563. John Hopkins University Press, Baltimore & London. Tyler M.J. and Davies M. (1986) Frogs of the Northern Territory. Conservation Commission of the Northern Territory, Darwin. Tyler M.J. and Davies M. (1979) A new species of cave dwelling frog from the Mitchell Plateau, Western Australia. Transactions of the Royal Society of South Australia 103, 149-153. Tyler M.J., Davies M. and Martin A.A. (1977) A new species of large green tree frog from northern Western Australia. Transactions of the Royal Society of South Australia 101, 133-138. Tyler M.J., Davies M. and Martin A.A. (1983) The frog fauna of the tiarkly Tableland, Northern Territory. 'Transactions of the Royal Society of South Australia 107, 237-242. Wightman G. and Andrews M. (1989) Plants of Northern Territory Monsoon Vine Forests. Conservation Commission of the Northern Territory', Darwin. The Green Tree Frog Utoria caerulea is active both on the ground and in trees. (Paul Horner) Advice to contributors The Northern Territory Naturalist publishes original papers concerned with any aspect of the natural history' of the Northern Territory' or northern Australia. Contributers need not be members of the NT Field Naturalists’ Club. Submissions are considered on the understanding that the content has not been published or accepted for publication elsewhere, and that copyright is retained by the NT Field Naturalists’ Club. All manuscripts are refereed, and the Editors reserve the right to modify manuscripts to eliminate ambiguity and repetition, or otherwise improve communication. Submitted papers may be in the form of Articles (up to 5000 words) describing the findings of original research; or Short Notes (up to c. 1500 words) summarising research results or describing new or unusual observations. Articles should include an Abstract, Introduction, Methods, Results and Discussion, but these headings are not required in Short Notes. Consideration will also be given to reviews, essays, species descriptions and identification notes where the subject is relevant and the contribution significantly increases current knowledge. Submitted manuscripts should be ty'ped, double-spaced and with generous margins. Authors are encouraged to submit manuscripts in IBM-compatible digital format, on CD or by email. These should be in Microsoft Word or rich-text (RTF) format. Figures and diagrams may be supplied digitally or as clear, black ink drawings on white paper. Good quality photographs (digital, transparencies or prints) that illustrate the main subject matter of the article are also welcome. Articles should be written to be intelligible to readers without a specialist knowledge of the subject. Authors should consult the most recent edition of the journal as a guide to lay'out including table format and headings. References should be listed alpabetically at the end of the article in the following form: Beehler B.M., Pratt T.K. and Zimmerman D.A. (1986) Birds of New Guinea. Princeton University Press, Newjersey. Begg R.J. (1988) Sandstone Antechinus. In The Australian Museum Complete Book of Australian Mammals (ed. R. Strahan), pp. 30-32. Angus & Robertson, Sydney'. Menkhorst K.A. and Wpinarski J.C.Z. (1992) Distribution of mammals in monsoon rainforests of the Northern Territory. Wildlife Research 19, 295-316. Manuscripts should be sent to: The NT Naturalist editorial committee, NT Field Naturalists' Club Inc., PO Box 39565, Winnellie NT 0821. Email: helen.larson@nt.gov.au Northern Territory Naturalist No. 18 November 2005 Articles New location records for some butterflies of the Top End and Kimberley regions Dona/d C. Franklin, Bruce Michael and Max Mace 1 Do predators affect the survival of Macariapallidata larvae? Implications for biological control of Mimosapigra in the NT Blair Grace 8 Threatened fishes of the Northern Territory Helen K. Larson 14 A survey of nocturnal reptiles of Robin Falls, Northern Territory: implications of Bufo marinus Lorrae J. McArthur and Jeanne E. Y oung 21 Distribution and natural history of the cryptic Chameleon Dragon Chelosania brunnecr. a review of records Colin R. Trainor 34 Short Notes Avian granivores consume flowers, not just seed, of the Top End Bamboo Bambusa ambemica Donald C. Franklin 45 A population count and ecological notes for the little-known terrestrial orchid Didymoplexis pallens Donald C. Franklin and Raelee Kerrigan 51 Birds of Lake Nongra and surrounding bushland. Northern Territory Martin Gole 54 Use of tree hollows bv the Green Tree Frog Litoria caerulea at East Point Reserve, Darwin S.J. Reynolds 61 Advice to contributors 68 Butterfly records Biological control of Mimosa pigra Threatened fishes Chameleon Dragon & other reptiles Orchids and bamboo Birds and tree frogs Contents inside back cover 17014 Uniprint NT 11.05 NJ