A SURVEY OF KITTI'S HOG-NOSED BAT, CRASEONYCTERIS THONGLONGYAI , IN KANCHANABURI PROVINCE, WESTERN THAILAND

Introduction

 The precarious status of Craseonycteris thonglongyai reported by Pye in February 1981, suggested that a more detailed survey of the Sai Yok area, using electronic bat detection equipment, should be made. The Association for the Conservation of Wildlife ( ACW ) in Bangkok arranged for me to visit the area for a period of five weeks from December 16, 1981 to January 18, 1982. Accommodation at Sai Yok was provided by the National Parks Division of the Royal Forestry Department of Thailand, and my expenses while in Thailand were provided by ACW . The People's Trust for Endangered Species were kind enough to pay my air fare and insurance, and I was lent bat detectors and ancillary equipment by Professor Pye of Queen Mary College, London.

 The primary purpose of the trip was to find and estimate the population of Craseonycteris . The echolocation sounds of this bat have been recorded and analysed by Pye, and are quite distinctive. It would therefore be possible to identify these bats without needing to catch them for primary identification. Observations on the presence and behaviour of other species of bats in this area were also to be made.

I stayed at Sai Yok National Park Headquarters for a period of three weeks from December 18 to January 9. Guides and porters were provided by the director of the park to take me to nine caves in the vicinity of the park headquarters. When no guides or porters were available I found and explored a further twelve caves in the area between the River Kwae Noi and the main road. A three day expedition was made to the Huai Bong Ti area, where four caves were examined. The director of Hat Ngiu village provided accommodation for myself, my guide and my porters.

 No bats were found in sixteen of the caves in the Sai Yok area although several of them had been reported to contain bats at the beginning of December 1981, when the area was visited by Stebbings (Pers comm).

 The size of the caves varied quite considerably, from a narrow crack in the rock which I could barely penetrate, to large caves with numerous passages and chambers. In the area between the river and the main road the caves were located in limestone outcrops, usually of limited size, while the caves at Huai Bong Ti and Tham Tong Su were located in more extensive limestone hills. A large number of small cracks and crevices were also investigated as far as possible, although they were never found to contain any bats.

 In addition to the exploration of caves, observations of hunting bats were made with the aid of a bat detector and a cassette recorder, using a frequency dividing circuit especially designed and built for this trip by myself. Mist nets were occasionally set in the vicinity of the park headquarters.

Craseonycteris thonglongyai

 When this bat was first discovered it was reported from several caves in the Sai Yok area. When Pye visited the area in February 1981 he found only a small number of the bats in one of the known caves and a group of no more than 200 bats at Tham Wang Phra which is located outside the borders of the park. When Stebbings visited the area in early December 1981 he could find bats in only one of the caves, Tham Wang Phra , and reported that their numbers had dwindled to only 20-30. This was apparently due to the cave having been inhabited by a Buddhist monk who had erected a bamboo screen across the front of the cave and had blocked a number of other exits.

 When I visited this cave, accompanied by Dr. Boonsong Lekagul ,the bamboo screen was still present across the mouth of the cave and there was no guard present. The bats occupied a small side chamber and there were always five or six bats in flight while we were in that chamber, although they did not leave the cave. Most of the bats either remained where they were or moved back over the surface of the rock, retreating into small holes and crevices. I estimated that there were at least five groups of eight to ten bats in addition to those in flight, a total population of around fifty bats.

 A second group of bats was found at Tham Tong Su , about four kilometers to the east of Sai Yok National Park, and well outside its borders. The cave had been used by a Buddhist monk and a bamboo platform with some of his belongings was still there. At the back of the main chamber there is a narrow opening about one meter in diameter and beyond this the cave winds its way back into the hill. This part of the cave is noticeably hotter and more humid than the main chamber and it is in this portion that the Craseonycteris were found. At nine sites within this cave there were small groups of bats flying, each group apparently distinct from its neighbours. In some cases, although not all, a group of roosting bats could be seen associated with the flying bats. In other cases they could sometimes be heard even when they were not seen. It appeared that each group of flying bats represented a small group of ten or more individuals; the total population within the cave would then be about 100 bats. Positive identification of these bats was aided by one bat that did not attempt to retreat into a crevice but remained hanging about a foot above our heads while being observed quite closely by the light of our torches.

 The third cave in which I found Craseonycteris is in the Huai Bong Ti area. Bain and Humphrey (1980) reported the presence of Craseonycteris in one cave near Huai Bong Ti , but gave no details of its location and it is not known whether the cave which I visited is the same one. The present cave does not have a name and is quite difficult to reach. It consists of a main chamber with a few small side chambers and a tall chimney which ascends out of sight. When I reached the cave my porters had already entered it and had caught a single specimen of Craseonycteris by hand. After making recordings of this bat it was released inside the cave. When I investigated the cave there were no Craseonycteris near where the first specimen had been found. A small group of them was seen high up in the chimney but when disturbed by torchlight they retreated further up the chimney until they were out of sight. A small group of four or five other bats also inhabited this cave. They could not be heard on the bat detector and would not permit very close observation but from their appearance and behaviour I believe that they were Megaderma spasma.

 A number of very small bats were observed and recorded hunting near the National Park Headquarters on a number of evenings. From their size and the nature of their echolocation signals it is possible that these were Craseonycteris, although I was never able to catch one of them for positive identification. The echolocation signals used by Craseonycteris inside a cave are short, constant frequency pulses at a frequency of 70 kHz, with weaker components at 35 kHz and 105 kHz. The repetition rate of the pulses is significantly higher than that of most other bats. The recordings made by Pye also include some pulses in which the frequency was swept through a narrow range around 70 kHz. The bats observed hunting would generally be using short pulses swept in frequency, with most of the energy between 70 kHz and 75 kHz. Some sound was detected at 35-38 kHz and 105-110 kHz. Pulse repetition rate and duration (as determined later from recordings made at the time) were similar to the rates and durations of sounds produced by Craseonycteris in the caves.

 The bats would first appear at about 18:30 hrs, hunting high around the tops of the teak and bamboo at a height of 20-30 ft. (7-10 m)and would gradually descend to a height of 10-15 ft. (3-5 m) as they hunted. This corresponds to the hunting behaviour of Craseonycteris described by Boonsong and McNeely (1977). The bats generally hunted in small groups of possibly three or four bats and were clearly catching flying insects and not gleaning. They were only found hunting in areas where there was an opening in the teak/bamboo canopy. They were never observed hunting within the forest itself, or in areas where the canopy was closed over the road. Neither were they found in the open areas between the edge of the teak forest and the main road. A small group would invariably be found hunting at the roundabout near the park headquarters, and on a number of evenings they were observed hunting at points along a 2 km section of the road through the plantation. By 18:50 hrs the number of bats and the number of 'catch buzzes' would decrease and between 19:00 hrs and 19:30 hrs only an occasional single bat would be heard, using sounds more characteristic of those used by Craseonycteris inside the caves. I estimate that there were about fifty of these bats hunting in the area, and it is likely that these may have come from either Tham Wang Phra or from Tham Tong Su, both of which are only four or five kilometers away. This would be within easy flying distance for a fast-flying bat such as Craseonycteris.

Taphozous melanopogon

 One colony of these bats was found in the Hat Ngiu area, in a cave opening onto the SSE face of Hoop Krathing mountain. A large number of these bats took to the air as we entered the cave, which made it very difficult to estimate their population, but it was thought to be between 100 and 200. Two of the bats were caught for identification and one of them was afterwards released. The other, which had been caught by hand by one of my porters, was unfortunately injured and was therefore killed and later preserved in alcohol. This specimen is now in the possession of Dr. Boonsong Lekagul.

 A much larger colony of nearly 1000 bats was also observed living in the wall cavity of the Central Post Office in Bangkok. At about 18:00 hrs these bats would leave the roost, departing in groups of about 100 bats at intervals of a few minutes. It was noticed that a number of bats were returning to the roost at about 18:30 hrs, before all the bats had left and it was thought that this might be due to the presence of young bats in the roost. When they left the roost the bats all departed in an easterly direction away from the river. By 18:30 hrs however, they could invariably be heard hunting near the ACW offices a few hundred meters to the south of the post office.

 Tham Khao Chong Pran, in Potharam district near Ratburi, was also found to contain several hundred T. melanopogon. This is a large cave which also contains a very large population of Tadarida plicata, estimated at three million by Boonsong. Most of the Taphozous lived in separate chambers from the Tadarida, although a small group was found in a hollow at the edge of the main Tadarida chamber. When the Tadarida emerged at dusk a few Taphozous would emerge with them but would fly off in a different direction from the stream of Tadarida . About fifteen minutes after the start of the Tadarida exit, the main body of the Taphozous came out of a different exit nearer to their main roosting chamber.

 The echolocation signals of these bats are short constant frequency pulses at 14-15 kHz and at 28-29 kHz. During interception 'catch buzzes' the 28-29 kHz component drops out and only the lower frequency is used. The 'catch buzzes' are frequently audible without the aid of a bat detector.

Rhinolophus malayanus

 Single individuals were heard in two caves in the Sai Yok area. In both cases they were heard flying after having been disturbed and their roosting location was not observed. In one of these caves they appeared to be co-habiting with a group of five or six Megaderma spasma.

In the other cave a solitary Rhinolophus robinsoni was found and a small group of five or six bats which were not positively identified but which could also have been R. malayanus. The latter cave was extremely large with many side chambers and the three groups of bats were each found on separate visits and in different parts of the cave.

 A single R. malayanus was also found at Tham Ta Ae in the Huai Bong Ti area. This cave consisted of a single passage with a small chamber at the far end. Another small passage extended beyond this chamber but was too small to be entered. The bat was found alone in the terminal chamber, hanging from the wall. It was undisturbed by my approach and I was able to take it by hand and carry it out of the cave without it waking up. The time from catching the bat to its awakening sufficiently to start echolocating was between five and ten minutes. It seems clear that this bat was in hibernation.

 On one evening a single R. malayanus was heard flying near the Sai Yok park headquarters, but it did not remain in the area. The echolocation sounds of this bat are characteristic of those used by all Rhinolophus so far studied. The pulses are long and of constant frequency, with a frequency downsweep at the end. The frequency of the c.f. portion is usually species specific and in this case was 90 kHz.

Rhinolophus robinsoni

  Several R. robinsoni were found in one of the caves in the Sai Yok area over a period of three days. On the first visit to the cave a single bat was found hanging from the roof of the main chamber in a fairly exposed position. This bat was not disturbed by my approach and I was able to lift it down from its roosting place and climb out of the cave with it before it awoke. It is clear that this bat, like the R. malayanus found near Huai Bong Ti , was in hibernation at the time. This specimen was kept in captivity for a period of one and a half days before being released in the forest near the park headquarters. It was noticed that in this specimen, a male with a forearm of 47 mm, the broadening of the sella continued to its base without a lower constriction.

 When this cave was revisited the following day, a female (forearm 50 mm) was seen flying in the cave and was caught with a hand net. During the transfer of this bat from the net to a cloth bag its cries produced a response from one of the side chambers of the cave. When this side chamber was investigated a further R. robinsoni was found which was torpid. On a third visit to the cave a day later no bats were to be found there.

 The bat which was caught with the handnet unfortunately died before it could be released and was therefore preserved in alcohol and is now in the possession of Dr Boonsong Lekagul .

 A further R. robinsoni was found in the large cave near the Burma railway. This bat was alone in a small side passage and it took considerable disturbance with a handnet to dislodge it. It is probable that this bat was also torpid. It was released in the cave after having been identified.

 Until 1972 this species was known from a single specimen from Southern Thailand. Since it has been shown that R. klossi is a subspecies of R. robinsoni the range has been extended to Malaya and to many of the east coast islands. The observations reported here extend the range of this bat for a considerable distance to the north.

 The echolocation sounds of this bat are typical of the genus with the c.f. portion at 60 kHz.

Rhinolophus sp.

 One further species of Rhinolophus was frequently observed hunting near the park headquarters. A single individual would appear each evening at 18:30 hrs and would then fly round the park office at a height of a few feet. It would occasionally leave its main beat and fly round the outbuildings or change its direction, but would generally return to its usual flight path. This behaviour would continue for about twenty minutes.

 A number of these bats would also be seen regularly near the roundabout. Shortly after 18:25 hrs they would be seen flying low through the forest, all apparently following the same path. On one evening eight bats were seen to cross the road within a few meters of the same spot, at intervals of a few minutes between 18:25 hrs and 18:30 hrs. A short while later at about 18:50 hrs a few of these bats would be heard hunting near the roundabout.

 The roosting site of these bats was never found and they were very difficult to catch in mist nets. On more than one occasion a bat was seen to fly up to a net and then turn and search along its length for a way past. A single specimen was caught but could not be identified from the key in 'The Mammals of Thailand' and is therefore being sent to John Edwards Hill at the British Museum for identification.

 The frequency of the echolocation sounds of this bat was 78 kHz. This is somewhat lower than that of R. malayanus although this bat is slightly smaller with a forearm of 40 mm.

Megaderma spasma

 A small group of bats co-habiting with Craseonycteris in a cave at Huai Bong Ti are believed to have been M. spasma. Two further groups were found in a cave in the Sai Yok area. This cave consisted of a long passage with a number of openings to the outside. In one of the terminal chambers a single Rhinolophus was heard, identified as R. malayanus by its echolocation frequency, and a group of Megaderma was found roosting. The Megaderma could not be heard on the bat detector and would not fly away until approached fairly closely. The bats remained in the chamber which was noticeably hotter and more humid than the rest of the cave. One of these bats was caught with a hand net and was later released in Bangkok .

 A second group of M. spasma also of five or six bats, was found in a different chamber some distance from the first. This chamber was also considerably hotter and more humid than the rest of the cave.

 The echolocation sounds of these bats are very short, broad band pulses, but they are generally too faint to be heard on a bat detector in the field.

Other bats

 Two caves containing Tadarida plicata were visited near Ratburi. Tham Rakang cave, in Muang district, was estimated by Dr Boonsong Lekagul to contain about one million bats, but is threatened by local limestone quarrying. The possibility of making this a protected area is being investigated by Forestry Department officials consulting with the First Army Engineers Department since Tham Rakang is on military ground.

 A large colony of T. plicata, numbering over three million bats according to Dr Boonsong, exists at Khao Chong Pran in Potharam district. In this case the cave itself has been grilled by local monks, but the bats are still threatened by hunters who set nets on the trees near the cave. As many as 10,000 bats may be killed in a single day. The bats are then sold to local restaurants since the bat meat and blood are considered to be potent aphrodisiacs.

 Both of these colonies provide a living for many local villagers who mine the guano and sell it to farmers, to whom it is a valuable source of fertiliser. At Tham Rakang twenty families are said to make a living from this trade and at Khao Chong Pran the sales earn at least 100,000 baht per year according to the monks whose land the cave is on. This pays for the repair and maintenance of the temple and also helped to pay for a school and hall within the temple compound.

 Tadarida plicata were also heard hunting every evening in the Sai Yok area. The echolocation sounds of these bats could be detected from 18:15 hrs in considerable numbers, especially in the more open and marshy areas near the main road, and occasional individuals could often be heard till late in the evening. The echolocation sounds of these bats are very loud, shallow frequency sweeps between 20 and 23 kHz. The bats may therefore be detected at a considerable distance and these hunting bats were never actually seen.

 At a third cave in the Ratburi area, Tham Khao Bin, two species of bat were found in considerable numbers, although they were so widely distributed in the extremely large cave that it was not possible to make an estimate of their population. The echolocation signals of these bats was typical of those used by Rhinolophus or Hipposideros at 68 kHz and 82 kHz. From the size of the bats they are believed to have been Rhinolophus but since they could not be caught precise identification is not possible. It is believed that the 68 kHz species was the same as that of a number of bats heard flying in the early evening near Hat Ngiu village.

Conclusions

 The protection of Craseonycteris thonglongyai is still of the utmost importance. The known population is still no more than it was estimated to be a year ago (although differently distributed) despite extensive searches using sophisticated electronic bat detectors. The main efforts should be directed towards Tham Tong Su . This site is known only to a limited number of people and contains a large proportion of the known population. Grilling of this cave would be feasible but should be approached with caution since it is not uncommon for snakes to use a grill as a perch from which to catch emerging bats. The fitting of a grill can also cause significant disturbance to the bats and advice from experts in this field should be sought before proceeding. If a grill is fitted it is important that it should be accompanied by clear notices explaining why it is there, so that there will be little motivation for people to break in in search of 'treasure'.

 The provision of physical protection for Tham Wang Phra is more difficult due to the size of the opening. At the beginning of December 1981, Dr Stebbings was assured by the Director of the National Parks Division that the bamboo fence had been removed and that a guard had been placed on the cave. However, when I visited the site, neither of these measures had been implemented. An important aspect of protection for this bat must be by widespread publicity and public education regarding its rarity and its scientific importance. The interpretive centre which is due to be built at Sai Yok National Park should play a key role. Precise location of the known roosting sites should however not be made generally available and efforts should be made to provide legislation protecting such sites when they lie outside the National Park. All visits to the occupied caves should be strictly vetted and controlled and such visits should be kept to a minimum.

 The protection of Craseonycteris must be considered in relation to the general conservation measures within the National Park. One of the most striking things about the park is the rate at which tourist amenities are being developed. Destruction of the natural environment for the benefit of man must be viewed with the gravest concern. A National Park should be primarily concerned with the preservation of the environment, not with the development of tourism. Its secondary function should be education of the public about natural history and the importance of environmental conservation. The establishment of an effective interpretive centre is of course fundamental to this and it should concentrate on factors relevant to Sai Yok and illustrate its lessons with examples that can be seen within the park. Thirdly the park should provide a clearing centre for scientific information about the park and its environs. At the least a day book should be kept in which any observations, whether made by Park rangers, visitors or visiting scientists should be entered. Even simple meterological observations can be useful. The habit of killing snakes within the Park is of course to be deplored, but even if this practice cannot be stopped then at least records of when and where any snakes were killed and of what type should be kept.

 Future expeditions looking for Craseonycteris should investigate the range of hills to the east of Sai Yok . Limestone hills such as these are liable to contain a great many caves which will be far deeper and more extensive than those near the river. The range of hills to the west, towards the Burmese border, are also promising but are somewhat inaccessible due to the density of the jungle in that area.

 I was often handicapped by lack of transport. When guides and porters were not available I was restricted to investigating caves in a limited area near the Sai Yok waterfall. Any further expeditions should have independent transport and should be capable of operating independently when necessary. It should be equipped with caving apparatus and include at least one experienced pot-holer. In my investigations I came across a number of caves or deep chambers that could not be explored due to the lack of suitable equipment

 The time of year is also of importance. While it is known that Craseonycteris lives in the area during the dry season, it appears to be a poor time for bats in general. There were very few bats hunting in the evenings and the disappearance of bats from at least one cave suggests that many bats leave the area at this time of year. It also appears that some species go into hibernation at these times, as evidenced by the torpid specimens of R. malayanus and R. robinsoni. There were very few nocturnal flying insects and even a mosquito net was generally unnecessary. Observations of Craseonycteris during the rainy season are lacking and an effort should be made to see if there is a significant change in the size of the known colonies at a time of year when there is a more plentiful food supply.