This section contains an assessment of the bat fauna on Buton Island with observations from a few other sites in the area (13.2), an analysis of the effect of wing morphology on community structure (13.3), and assessment of the small mammal fauna of Buton Island (13.4).

The aim of the present study was to produce a preliminary description of the bat fauna of South East Sulawesi. Efforts focused on the bat community in the Kakenauwe Forest Reserve and surrounding habitats (near Labundo bundo, Buton Island), with more limited evaluations of species representation in Rawa Aopa National Park (mainland S.E. Sulawesi), Kaisabu Cave (Buton Island), Kabaena Island and Hoga Island. Diverse methods were employed (harp traps, mist nets, cave and roost visits) to capture bat species. In excess of 600 individuals and 33 species were captured in the period from 07 July to 15 September 2000, constituting nearly half of the species expected to occur in Sulawesi. The bat community at Kakenauwe comprised 29 species and included four endemic species, five restricted range species and two possible new records for Sulawesi. The fauna recorded at Rawa Aopa was a subset of that found in Kakenauwe, with the exception of a potentially undescribed species of Hipposideros. Kaisabu Cave is an important roost site for eight species, including one endemic and three restricted range species; and early results from Kabaena suggest that caves are of considerable importance to the Island’s bat fauna. Ecomorphological data include wing morphology and echolocation signal design and are to be used to assess evidence for community structure at Kakenauwe.

Introduction

Despite the biogeographic significance of Sulawesi within the Wallacea region, the bat fauna remains poorly known. In particular, there have been no surveys of bats of Buton Island prior to the present study. Thus the primary aim of the present study was to characterize as fully as possible the bat fauna of Buton Island. Species discovery rates for bats are such that efforts focused primarily on the community in and around the Kakenauwe Forest Reserve near Labundo bundo.

To maximize the information derived from a relatively simple sampling program, ecomorphological data that might clarify species identification and elucidate community structure were also collected. Specifically, echolocation signal design and wing morphology are to be described for each species.

Bat Sp. Photo Tigga Kingston

Species differ in their susceptibility to different capture techniques. Four-bank harp traps were deployed across forest trails to sample the insectivorous forest interior guild. Up to four traps were used per night, spaced a minimum of 50 m apart. Trap positions were changed each night. Traps were deployed from 6 pm; attended regularly until 7.30pm and then left over night to be checked at dawn before closing. A trap night thus constituted the 12 hour period from 6pm to 6am. Mist nets were used to capture fruit bats, and were set in the village of Labundo bundo and the surrounding gardens. Mist nets set over rivers were also the primary means by which the aerial insectivorous bat guild was sampled. Flick netting was occasionally attempted to catch small insectivorous bats flying in Labundo-bundo.

Mist nets and harp traps were used during cave visits, in combination with hand nets.

Species were identified following Corbet and Hill (1992), and Payne and Francis (1985). All individuals were sexed, reproductive condition assessed, weighed, and length of forearm measured. Voucher specimens for each species were taken, to be lodged at the Bogor Museum. Echolocation calls were sampled using an S25 bat detector linked to an S350

sound processor (time expansion x 10) (Ultra Sound Advice) and downloaded to a Professional Sony Walkman. Hipposiderid and rhinolophid bats were recorded in the hand (resting frequency); narrow-space insectivorous species (Kerivoulinae and Murininae) were recorded in a flight cage; and aerial insectivorous bats (Vespetilionidae, Molossidae) were recorded free-flying on release. Wing tracings were made for all insectivorous species: the individual was placed face down on a piece of paper, the right wing extended perpendicular to the anterior-posterior body axis and traced.

33 species were recorded in the course of the present study. Many species exhibited divergence from current species descriptions; definitive identification requires examination of cranial features and comparison with existing collections held at the American Museum of Natural History. All identifications listed below are preliminary and subject to confirmation.

1. Kakenauwe Forest Reserve and Labundo bundo.

7 Jul - 29 Jul 00; 14 Aug - 28 Aug 00; 04 Sep - 09 Sep 00

29 species in 80 harp trap nights and 30 mistnet . Records of particular interest:

Rhinolophus celebensis -- restricted range (Sulawesi, Java, Bali + Islands)

Rhinolophus philippinensis -- large and small morphs occurring syntopically

Hipposideros dinops -- Sulawesi endemic

Hipposideros cineraceus ? -- if confirmed is a new record for Sulawesi

Hipposideros galeritus ? -- if confirmed is a new record for Sulawesi

Miniopterus tristis – restricted range (Sulawesi, Philippines, New Guinea)

Tadarida sarasinorum -- restricted range (Sulawesi, Mindanao (Philippines), Peleng Island)

Cheiromeles parvidens – restricted range (Sulawesi, ?Mindoro (Philippines))

Dobsonia viridis – restricted range (Sulawesi and Moluccas)

Rousettus celebensis -- Sulawesi endemic

Boneia bidens – Sulawesi endemic

Thoopterus nigrescens -- Sulawesi endemic

1 Aug - 10 Aug 00

32 harp trap nights. 12 species.

Bat species at Rawa Aopa were a subset of the forest interior guild found at Kakenauwe with the exception of a currently unidentified Hipposideros sp.

3. Kaisabu Cave

30 Aug - 3 Sep 00

Hand netting, mistnetting, harp trapping.

Eight species were recorded from this cave: Miniopteris australis, M. tristis (restricted range), Emballonura sp, Hipposideros cervinus, Rhinolophus arcuatus, R. celebensis (restricted range), Dobsonia viridis (restricted range), Rousettus celebensis (endemic). A full report is in preparation, but the cave is clearly an important roosting site for a number of restricted range/endemic species and it is strongly advocated that further visits be restricted to those for scientific purposes only. Occupation rates declined markedly in the course of the present survey, indicating that these species are highly sensitive to disturbance.

05 Sep - 09 Sep 00

Preliminary survey undertaken by Nick Berry and Mairi Pierce

Four caves were visited, all of which were occupied by a mixture of fruit bats and insectivorous bats (including a large Hipposideros sp. tentatively identified as the Sulawesi endemic H. inexpectatus).

5. Hoga Island

12 Sep - 16 Sep 00

10 Mist net hours

Very limited sampling with mist nets confirmed the presence of four Megchiroptera species on the Island: Pteropus hypomelanus, Dobsonia viridis, Rousettus celebensis and Macroglossus minimus.

Anticipated publications and completion dates. Titles are tentative and contingent upon the results of further analysis.  Note most of these publications have been delayed because the Indonesian Institute of Sciences has to date not released the various specimens collected and currently in the Bogor Museum in Java. 

First record of Hipposideros cineraceus and H.galeritus from Sulawesi (March 2001)

A new Hipposideros species from South East Sulawesi (May 2001)

An annotated checklist of the bats of Buton Island (June 2001)

Niche packing or spreading? Species richness and community structure in bat communities from South East Asia (Oct 2001).

Social structure in the Tube-Nosed Insectivorous Bat (Murina florium); genetic and behavioural evidence for monogamy. (Nov 2001)

Echolocation signal design in bats from Sulawesi (Dec 2001)

Call frequency and prey size in five species of insectivorous bats from South East Asia: a test of acoustic resource partitioning. (Dec 2001)

The relationship between wing morphology and flight performance has long been recognised. Wing parameters of particular importance are aspect ratio (an index of wing shape defined as the wing span ^2 / wing area) and wing loading (mass x gravity/wing area). A high aspect ratio describes a long narrow wing and maximizes flight efficiency at the cost of manoeuvrability. Manoeuvrability is enhanced by low wing loading values, but flight is consequently slow and energetically costly. Different combinations of these parameters characterize different guilds. For example, bats that forage for dispersed prey items in the open spaces above the forest have high wing loading and aspect ratios facilitating fast, efficient flight. Species foraging within the forest interior face the challenge of locating and pursuing prey in a cluttered environment. Thus the primary selection pressure on narrow space insectivores is to be manoeuvrable and most species are characterized by low wing loading and aspect ratio as a consequence. The aims of the present study were three fold. First, to test the correspondence between wing morphology and insectivorous guild membership. Second, to determine the relationship between clutter tolerance and wing morphology for the narrow space insectivorous guild of the forest interior. Third, to further validate this relationship by direct performance testing.

Wing tracings were made as described above (section 13.2). Ten individuals of each sex of each species were drawn when species representation permitted. From these tracings, aspect ratio and wing loading can be derived. Canonical discriminant analyses will test the correspondence between guild allocation and wing morphology. Within the forest interior, the degree of clutter at each harp trap position was characterized by recording the percentage visibility of a 3m marked pole. Visibility was assessed at 5 m and 10 m from the trap along transects that entered the forest at 45 degrees from the trap in both directions and on either side of the trail (resulting in 4 measures per trap). Canopy cover was recorded photographically at each trap postion. The relationship between local clutter, species representation and wing morphology is to be tested. A subset of the forest interior guild, primarily species belonging to the vespertilionid subfamilies Kerivoulinae and Murininae, were used in the flight performance tests. Individuals were flown in a flight cage and required to negotiate two banks of vertical strings. The banks were approximately 1m apart, and the distance between the strings was set at 10 cm. Individuals were scored on their ability to negotiate the banks of strings and the mean score of 10 passes calculated.

Wing parameters will be derived from the tracings in the course of the academic year. Preliminary results of the flight trials suggest that performance is not size-related; the smallest species were not necessarily the most adept at negotiating the artificial clutter. This indicates that size-independent differences in wing morphology may have a role in structuring communities and be a mechanism by which resource partitioning is effected.

A dissertation entitled Wing morphology and community structure in an insectivorous bat community from Sulawesi was produced in March 2001 by Rachel Watts from the University of Glasgow.

Introduction

The biodiversity assessment of the small mammal fauna was carried out over a 2 month period beginning in July 2000. Trapping was concentrated around the forests of central Buton and a short visit was also made to the Rawa Aopa National Park on mainland SE Sulawesi. Five trap lines spaced approximately 50m apart along a forest trail were used. Each trap line comprised 10 Longworth traps and 4 tomahawk traps, and was usually run for five nights.

Processing a Malay Civet under anaesthetic ; Photo Andy Jennings

Every animal captured was anaesthesised with Ketamine Hydrochloride. A sample of ear tissue was taken as a DNA sample and the following body measurements and descriptions were taken: weight, sex, reproductive condition, head and body length, tail length, hind foot length, and pelage description. Photographs were taken and the animal was fur clipped before it was released unharmed. Over 300 rodents were captured (including recaptures) and over 200 DNA samples were taken. Although there was a good capture rates of rat-like animals, no squirrels were captured and the trapping technique needs to be modified to include this group in future programmes. Three species of shrews were caught in pitfall traps. Identification of each animal has proved very difficult as there is very little published information, but it would appear that there are possibly 10 species of rodents within the areas that were trapped. Further analysis of the measurement data and DNA samples taken should clarify the identification situation. At least one type specimen has been taken of each different species for future skull measurement comparison with known museum collections. To investigate possible microhabitat preferences of each species, habitat plots around each trap site were set up and various habitat variables recorded. A multiple regression analysis will be used to analyse this data. In addition to the small mammal trapping two large wire cage traps were set in various locations around the Labundo Bundo base. The aim of this exercise was to determine how to trap civets for a possible radio-tracking project next year. Traps were placed along large logs, or at the base of large trees, and baited with cat food. After 21 trap nights 3 male Malay civets were caught. An attempt to process these animals using Ketamine was made, but unfortunately the potency of the drug had diminished from exposure to high daytime temperatures and the animals did not respond fully. However, it was possible to weigh some of the animals and measure their neck circumference. Two male Malay civets were also caught in the small Tomahawk traps that were set for rats and squirrels.

A dissertation entitled Micro-habitat Preferences of the Rodent Fauna of Buton Island was produced in March 2001 by Niall Young from Sparsholt College, Hampshire.  This dissertation is available at the Labundo field centre and from the Op Wall UK offices.