Tragulus Javanicus)

Mammal Study 31: 111–114 (2006) © the Mammalogical Society of Japan Short communication

Social system of the lesser mouse-deer (Tragulus javanicus)

Hisashi Matsubayashi1,*, Edwin Bosi2 and Shiro Kohshima3

1 Faculty of Agriculture, Tokyo University of Agriculture, 1737 Funako, Atsugi, Kanagawa 243-0034, Japan 2 University of Malaysia Sabah, Locked Bag No. 2073, Kota Kinabalu, Sabah, Malaysia 3 Tokyo Institute of Technology, Meguro-Ku, Tokyo 152-8551, Japan

The lesser mouse-deer (Tragulus javanicus), the smallest mean area of error polygons was 29.9 ± 19.9 m2 (n = 10). ungulate species in the world (Ronald 1991), is a com- Indeed, dropped transmitters were located within about mon native to the tropical forests of Southeast Asia. As 10 m from the estimated location. Precise methods of the mouse-deer is a primitive ruminant, investigating the field mapping and animal trapping conducted in this their social system is necessary for understanding the study were reported in our previous paper (Matsubayashi evolution of social systems in ungulates. Except for a et al. 2003). The size of the home range and the core few field reports on group size (Ratnam 1982; Miura and area were determined with minimum convex polygons Idris 1999) and on the fighting and marking behaviours (MCP) (Mohr 1947) and adaptive kernel techniques of males based on fragmental observations (Davison (Software: The Home Ranger version 1.5). For analyses 1980), there are almost no reports on their social behav- of the core-area overlap, we analysed the data of 6 indi- ior in their natural habitant. To study social system of viduals (3 males: M06, M10, and M12; 3 females: F01, the lesser mouse-deer in natural habitat, we analyzed 1) F13, and F14). Daytime core-areas, which included 50% the group size, 2) the core-area overlap, and 3) inter- of the locations recorded during the day, were calculated action among neighboring individuals. for individuals with >70 locations (Table 1). We used the data of daytime core-area analysis because the active Materials and methods behaviors of males were observed between 8:00 h and 10:00 h (Davison 1980), and our previous study also We conducted field studies at the Kabili-Sepilok For- found the lesser mouse-deer to be diurnal rather than est Reserve (5°51'N, 117°56'E), Sabah, Malaysia (Fig. nocturnal (Matsubayashi et al. 2003; Matsubayashi and 1). The study area was about 7 ha in size, located in a Sukor 2005). In order to investigate their courtship mixed dipterocarp forest in lowlands. The study period behaviour, we tracked a whole-day behaviour through 24 was over two years between May 1998 and March 2001. hour radio-tracking (every 2 hours, 13 trials in total). To We captured 15 lesser mouse-deer including 9 males compare the average distance between the pair, the dis- (8 adults), 6 females (5 adults), and conducted a radio- tance during courtship (13 locations in a day) and non- tracking study fitting radio transmitters (144 MHz, 11 g, courtship (79 locations over 36 days) were calculated. Holohil Systems Ltd., Ontario, Canada) to 11 individuals The size of the home range and core area were statisti- (7 adult males and 4 adult females). Locations of mouse- cally examined with t-tests. All tests were 2-tailed and deer were determined by measuring direction of the radio the alpha level was set as 5%. Data are presented as signal from >2 points. Locations of the animals were mean ± standard deviation. plotted on a 1 : 1,000 map. We tracked animals at dis- tances <100 m as beyond that we could not receive clear Results and Discussion transmitter signals. Therefore, location error was com- paratively small. For example, for one individual located During the study period, 70 individuals were observed 13 times in a full day of radio-tracking on 22 and 23 over a total of 66 times, 93.9% (n = 62) were solitary November 2000, the mean distance between animal and individuals, 6.1% (n = 4) were in a group with 2 individ- observer was estimated to be 30.3 ± 6.1 m (n = 30) and uals, and none was in a group with more than 3 individu-

*To whom correspondence should be addressed. E-mail: [email protected] 112 Mammal Study 31 (2006)

Fig. 1. Location of Kabili-Sepilok Forest Reserve, Sabah, Malaysia.

Table 1. Size of home-range and core area of the lesser mouse-deer in Kabili-Sepilok Forest Reserve.

Home-range size (ha) ID No. Tracking Period No. of location Minimum Core area of daytime Convex Polygon (50% Adaptive-kernel) Male 06 23 Mar 1999–22 Oct 1999 125 6.2 1.6 10 23 Jun 1999–27 Nov 1999 126 5.5 0.8 11 29 Oct 1999–21 Jun 2000 133 6.0 1.5 Mean ± SD 5.9 ± 0.4 1.3 ± 0.4 Female 01 3 Nov 1998–29 Nov 1999 263 4.7 1.2 13 2 Nov 1999–13 Jul 2000 72 3.5 1.1 14 10 Mar 2000–4 Dec 2000 143 4.9 1.3 Mean ± SD 4.4 ± 0.8 1.2 ± 0.1 als. The mean group size (including solitary individuals) instance, the core areas between neighbor females (F13 was 1.06. This result shows that the lesser mouse-deer is and F14) were completely separate, though the MCP highly solitary. home-range overlaps were 26.5% and 18.4% respec- Table 1 shows the size of MCP home-range and core tively (Fig. 2a). The core areas of neighbor males (M06 area of the lesser mouse-deer. Mean size of MCP home- and M10) were also completely separate, though their range tended to be larger in males than in females (P = MCP home ranges overlapped widely, 28.9% and 0.051), while mean size of core area was not signifi- 43.5%, respectively (Fig. 2b). In contrast, core areas of cantly different between males and females (P = 0.74). pair individuals were highly overlapped (M11 and F13: Fig. 2 shows that the core areas of neighbors were 64.9% and 76.3%, respectively) with widely overlapping completely separate among same-sex individuals but home ranges. The results suggest social system of the overlapped widely among opposite-sex individuals. For lesser mouse-deer is monogamy. In addition, an obser- Matsubayashi et al., Social system of lesser mouse-deer 113

Fig. 3. The movement of an adult male (M06: ▲) and an adult female (F01: ●) during the period when they were together (10:00 h 9 October–10:00 h 10 October 1999). The serial number shows the time at each location. Open square indicates resting sites in the night- time. Open circle indicates the place where they approached within 20 m. Arrows indicate the route of their movements.

ment together was 9.9 ± 5.0 m (n = 12), while that of non-movement together was 64.4 ± 44.8 m (n = 79). However, male of lesser mouse-deer also can be polygamous. Because 1) the core area of paired female Fig. 2. Relationship of the core areas among neighboring individu- als. a: Home ranges and core areas of one male (M11) and two (F01) overlapped not only with the core area of the females (F13 and F14) and M11 and F13, tracked over 3 months at the paired male (M06) but also with another neighboring same time. b: Home ranges and core areas of two males (M06 and male (M10) (Fig. 2b). This result suggests that males M10) and one female (F01) and M06 and F01, tracked over 7 months admit the invasion of females into their core areas and at the same time. Solid lines indicate the male’s core area. Broken lines indicate female’s core area. Painted areas indicate core areas of paired males do not control the movement of the paired the same-sex individuals (F13 and F14; M06 and M10). Polygon indi- female into the home range of other males. 2) We also cates MCP home-range. confirmed that neighboring males (M10 and M11) invaded the home range of a missing male (M06) that vation of a consort pair, M06 and F01, also supports their had moved out of its home range after mating, and monogamous social system. Fig. 3 shows the movement approached its paired female (F01). 3) Territory of of an adult male (M06) and an adult female (F01) with males is more temporary than that of females. The overlapping core area (64.6% for the male and 50.8% tracking term of males (less than 1–8 months, mean for the female) monitored by full-day radio-tracking = 4.3 months, n = 3) was much shorter than that of fe- while they were estimated to move together (10:00 h, 9 males (8–13 months, mean = 10 months, n = 3, Table 1). October–10:00 h, 10 October 1999). The mean distance Though the home range of this species has been believed between these individuals during the period of move- to be permanent (Klaus 1990), data of other males and 114 Mammal Study 31 (2006) females were eliminated because of problems with their References transmitters. These results suggest that the social system Barrette, C. 1987. The comparative behavior and ecology of chevro- of the lesser mouse-deer is facultative monogamy/ tains, musk deer, and morphologically conservative deer. In (C. polygyny (Clutton-Brock 1989). This social system is W. Wemmer, ed.) Biology and Management of the Cervidae. also found among small ungulates dwelling in the forests Pp. 200–213. Smithsonian Institution Press, Washington, D.C. and ecotone habitats of Africa such as duiker and dikdik Clutton-Brock, T. H. 1989. Mammalian mating systems. Proceedings of the Royal Society of London, series B 236: 339–372. (Jarman 1974; Barrette 1987). These species are more Davison, G. W. H. 1980. Territorial fighting by lesser mouse-deer. primitive and have simpler social systems compared to Malayan Nature Journal 34: 1–6. more developed larger ungulates dwelling in forests, ec- Jarman, P. J. 1974. The social organization of antelope in relation to their ecology. Behaviour 48: 215–267. otone, or open habitats. This result suggests that the so- Klaus, R. 1990. Chevrotains. In (S. P. Parker, ed.) Grzimek’s En- cial system of ungulates might relate to their phylogeny cyclopedia of Mammals. Volume 5. Pp. 120–123. McGraw-Hill as well as feeding style and group size (Jarman 1974). Publishing Company. Further field studies on other small ungulates dwelling Matsubayashi, H., Bosi, E. and Kohshima, S. 2003. Activity and habitat use of lesser mouse-deer (Tragulus javanicus). Journal of forests especially tragulids and small cervids will be nec- Mammalogy 84: 234–242. essary to understand the evolution of the social system of Matsubayashi, H. and Sukor, J. R. A. 2005. Activity and habitat use the ungulates. of two sympatric mouse-deer species, Tragulus javanicus and Tragulus napu, in Sabah, Malaysia, Borneo. Malayan Nature Journal 57: 235–241. Acknowledgments: We thank E. Tambing, Sabah Wild- Miura, S. and Idris, A. H. 1999. Present status and group size of the life Department for supporting our fieldwork. Our grati- mouse-deer on Pulau Tioman, Malaysia. Malayan Nature Journal tude goes to Y. Matsubayashi and anonymous referees 53: 335–339. Mohr, C. O. 1947. Table of equivalent populations of North American for useful suggestions on this manuscript. small mammals. American Midland Naturalist 37: 223–249. Ratnam, C. L. 1982. Preliminary observation on the Tragulidae at Kuala Lumpur. Journal of Wildlife Parks 1: 29–34. Ronald, M. N. 1991. Walker’s Mammals of the World. 5th ed. Pp. 1360–1362. The Johns Hopkins University Press, Baltimore and London.

Received 26 January 2006. Accepted 7 June 2006.