(Eulemur macaco macaco) at Ampasikely, northwest Madagascar

American Journal of Primatology 67:299–312 (2005)

RESEARCH ARTICLE Demography, Range Use, and Behavior in Black Lemurs (Eulemur macaco macaco) at Ampasikely, Northwest Madagascar FRANC ¸ OISE BAYART1n and BRUNO SIMMEN2 1 De´partement Ećologie et Gestion de la Biodiversite´, CNRS-UMR5176, Museé National d’Histoire Naturelle, Brunoy, France 2 De´partement Hommes, Natures et Socie´te´s, CNRS-UMR5145, Museúm National d’Histoire Naturelle, Brunoy, France

We studied a black lemur population over a 2-year period (1992–1993) and 8 years later (2000) in a 50-ha secondary forest in northwest Madagascar. All of the animals were marked to investigate population dynamics and seasonal variation in ranging and behavior, and new data on black lemurs were obtained. Our data on demographic characteristics were expanded to include other forest sites and contrasted with those collected in other Eulemur macaco macaco field studies, in relation to human activity and the presence of introduced and cultivated plant species. Density is affected by deforestation and hunting. Group size and home range depend on the composition of the forest and probably food patches. Sex ratio at birth varies according to the number of females per group, a result that fits the local resource competition model. Groups are multimale-multifemale, and adult females form the core of the groups. Reproductive parameters indicate sharply defined seasonal breeding, a high female reproductive rate, and birth synchrony. Changes in group composition reveal male and female juvenile dispersal, male transfer between groups at the time of mating, and adult female transfer and group fission when groups exceed a critical size. At mating and birth, intergroup agonistic encounters occurred at home-range boundaries, and larger groups were dominant over smaller groups. Patterns of intragroup interactions suggest that males compete for access to groups of females during the mating season, and that females may compete for food resources during the birth season. Our study also reports female social dominance and lack of sexual weight dimorphism in this species. Am. J. Primatol. 67:299–312, 2005. r 2005 Wiley-Liss, Inc. Contract grant sponsor: CNRS; Contract grant sponsor: Institute of Embryology of the University Louis Pasteur; Contract grant sponsor: French Ministry of the Environment (ECOFOR/MNHN); Contract grant number: 2000.18. n ´ cologie Geńe´rale, Museé Correspondence to: Françoise Bayart, CNRS-UMR 5176, Laboratoire d’E ˆteau, 91800 Brunoy, France. National d’Histoire Naturelle, 4 Avenue du Petit Cha E-mail: [email protected]

Received 11 July 2002; revised 22 March 2005; revision accepted 22 March 2005 DOI: 10.1002/ajp.20186 Published online in Wiley InterScience (www.interscience.wiley.com).

r

2005 Wiley-Liss, Inc.

300 / Bayart and Simmen

Key words: Eulemur macaco; demography; ranging patterns; social structure; seasonal variation

INTRODUCTION The life history traits of group-living lemurs are now recognized to be very different from those of most other group-living primates. Such traits include fewer females per group on average, equal adult sex ratios, male and female transfer between groups, sharply seasonal breeding, lack of sexual size dimorphism, female aggression, strong male–female bonds, and female feeding priority over males [Jolly, 1998; Kappeler, 1997, 1999, 2000; Kappeler & Ganzhorn, 1994]. After the ‘‘female need hypothesis’’ was proposed by Jolly [1984], the ‘‘energy conservation hypothesis’’ [Pereira, 1993; Pereira et al., 1999] and ‘‘energy frugality hypothesis’’ [Wright, 1999] emphasized how the ecological constraints of a harsh and unpredictable environment in Madagascar have shaped the evolution of major lemur traits. Elsewhere, van Schaik and Kappeler [1993, 1996] focused on the role of the Holocene extinction of large diurnal raptors in the emergence of lemur cathemerality (i.e., the transition from a nocturnal to a diurnal lifestyle), as well as small group size in non-nocturnalliving species (i.e., evolving from pair-living). To understand lemur adaptations, it is useful to compare different populations of the same species [Sussman, 2002]. As regards the Lemuridae, studies on demography, ranging, social behavior, and seasonality have focused on Lemur catta [Gould et al., 2003; Jolly et al., 2002] and Varecia [Vasey, 2003], but data are sparse for most Eulemur species [Overdorff & Johnson, 2003], except for Eulemur fulvus [Overdorff et al., 1999; Tattersall, 1977]. In this study we investigated the socioecological mechanisms by which the demography of black lemur populations (Eulemur macaco macaco) is regulated by comparing data collected in three locations in northwestern Madagascar in 1992–1993 and 2000 with those collected by other researchers in other northwestern forests in recent years. Toward that end we sought to determine the limits of variation in density, home range, group size, and group composition in populations inhabiting different habitats. Long-term data on a population of black lemurs in northwest Madagascar allowed us to analyze demographic changes along with major ecological and reproductive correlates. Our main study site (Ampasikely, 131250 131260 S, 481280 -481290 E; Fig. 1) was a secondary forest (50 ha) with old plantations and introduced plant species. Additionally, in 1992 we assessed population densities from group censuses at two other study sites, Ampangorina (north shore of Nosy Komba; 131260 3000 S, 481210 E) and Ambalahonko (south shore of Nosy Be, 131240 S, 481210 E, just outside the boundaries of Lokobe Reserve). Ampangorina is a village surrounded by various crop fields and degraded secondary forest (50 ha), where black lemurs are protected by the villagers and are fed bananas on a regular basis. Ambalahonko and its surroundings (200 ha) are secondary forest composed of mangroves, extensive bare ground areas due to deforestation, and relics of primary forest where hunting is prohibited. Other black lemur populations living nearby have also been studied, at Ambato Massif in ‘‘well established secondary forest’’ [Colquhoun, 1993, 1998a], and in the primary forest of Lokobe Reserve on Nosy Be [Andrews & Birkinshaw, 1998; Birkinshaw, 1999]. Since those studies and ours were conducted simultaneously, it is feasible to compare these sites in terms of black lemur demography, range use, and behavior.

Am. J. Primatol. DOI 10.1002/ajp

Black Lemur Demography and Behavior / 301

Main roads Reserve boundaries

Scale

N

NOSY BE

10 km

0

NOSY FALY LOKOBE NATURE RESERVE

HELL-VILLE

NOSY FALY PENINSULA Ambalahonko Ampangorina Ampasikely

NOSY KOMBA

Antsatsaka AMBATO MASSIF

Ambaliha

ANKIFY PENINSULA

AMBANJA

AMPASINDAVA PENINSULA

Beraty

MANONGARIVO SPECIAL RESERVE

MAROMANDIA Fig. 1. Black lemur study-site locations in the northwest of Madagascar (map derived from Andrews (unpublished results)). Ambalahonko: fragmented secondary forest. Ampangorina: cultivated secondary forest with provisioning of the lemurs. Ampasikely: secondary forest with cultivated and introduced plant species.

MATERIALS AND METHODS

Study Site The main study site (Ampasikely) is a private landholding located along the west coast of the Nosy Faly Peninsula. In addition to mangroves, the area is



composed of relics of primary forest and of dense, low-canopy, seasonally moist, semideciduous secondary forest. The flora [Simmen et al., in press] include many species that differ from those found in the Lokobe rain forest [Birkinshaw, 1999] or Ambato Massif secondary forest [Colquhoun, 1993]. Cultivated plants are particularly important since they form large clumps of food for the lemurs, and include cashew (Anacardium occidentale; 3 ha), mango (Mangifera indica; 2 ha), Albizia lebbeck and Piper sp. (2 ha), Coffea sp. (1 ha), and Carica papaya (1 ha). Annona spp., Artocarpus spp., Musa spp., Citrus spp., Ceiba pentandra, Cananga odorata, and Albizia saman are grown near local houses. The rainfall and temperatures that were observed during our study [see Colquhoun, 1993] are typical of the Sambirano climate, with an austral summer from November to April (rainy season) and an austral winter from May to October (dry season). The population has suffered from sporadic hunting. Since 1992 an effort to prevent hunting has been made, concomitant with the construction of a hotel on the seashore. The nearest black lemur population was found 5 km to the southeast of our study site. The only other resident prosimian species is the nocturnal Mirza cocquereli (Cheirogaleidae). Potential predators were raptors, whose sighting was responded to with alarm calls by the lemurs (see also Colquhoun [1993]). There were no constrictor snakes or large carnivores (except dogs).

Captures and Behavioral Observations At this site, the black lemur population was censused in May 1987 by Meier and Rumpler [1992], who earmarked the individuals in 1988 by Andrews (unpublished results), and in 1992–1993 by Bayart et al. [1993] and Rabarivola et al. [1998]. In January 1992, 29 out of 30 individuals were captured (with the use of a blowpipe and ketalar anesthetic (0.3 ml/kg)), weighed, measured, bloodsampled, and marked with collars. They were then released at the site of capture. We put collars on adults (eight males and eight females) and juveniles (three males and three females). Infants (five males and two females) were simply earmarked and tail-marked for identification. Infants born in 1991 and juveniles born in 1990 were recaptured in May 1992 so that they could be collared. Infants born in 1992 (four males and four females) were captured in April 1993, when individuals born in 1990 were recaptured. In 1992–1993, the population was studied for a total of 7 months 3 weeks across different seasons (Table I). In 2000, we returned to the site for 3 weeks to assess survival and gather additional data to complement our 1992–1993 data set. We used ad libitum observations to map group locations, using trails with trees marked at 15-m intervals, and to record group composition and the

TABLE I. Observation Protocol

Reproductive phase 1992 1992 1992 1993 2000 a

Season

Dates

weaning season Middle rainy Jan 17 to Feb 23 mating season Middle dry May 24 to Aug 3 birth season Late dry Sept 4 to Oct 6 mating season Early dry March 26 to June 23 birth season Late dry Sept 24 to Oct 15

Daysa Scan-days 30 42 21 76 21

10 15 18 28 15 2

Including scanned and ad libitum observations; (D): daytime; (N): nighttime observations.


5-min scans (hours) 240 (20, D) 540 (45, D) 324 (27, D) 1020 (85, D) 543 (45, D) 129 (11, N)


exclusion of individuals (i.e., animals seen alone at the same location for several days). A 5-min scan sampling method [Altmann, 1974; Crook & Aldrich-Blake, 1968] was used to record diurnal activity on focal animals from 5 a.m. until 7 p.m. In 2000, we also made observations throughout two nights of full moon. Morning and afternoon observation sessions were alternated among groups and distributed across different times of the day. The behavioral categories included travel (locomotion from one place to another); rest (alone or in contact); feed (or forage); affiliative intragroup interactions (allogrooming, allomarking, courtship, and mating); agonistic intragroup interactions (displacement, threat, slapping, and targeted aggression, as defined by Vick and Pereira [1989]); long-distance intergroup interactions (loud calls exchanged at a distance between groups); pacific intergroup interactions (at a short distance with affiliative calls); intergroup displays (marking, tail swishing, and leaping back and forth accompanied by loud calls); and intergroup chases (pursuing targeted individuals with loud calls and staccato grunts). For each period, all activities were analyzed as a percentage of scans. Pearson chi-square tests were performed to compare behavioral activities across the different reproductive periods. For inter- and intragroup interaction analyses, the two mating seasons (which did not differ from each other) were lumped together. RESULTS

Population Demography At the onset of the study (January 1992), only seven individuals (five adult males and two adult females) of the 50 captured in May 1987 were still present, and the population had been reduced to 30 individuals. The villagers explained this 40% decrease in the population within a 5-year period as the result of sporadic hunting. Between 1992 and 1993, the total population size fluctuated between 29 and 40 individuals (Table II). The population comprised three social groups (GP, GV, and GT) in February 1992, and four in June 1993. Overall, group size ranged between four and 14 individuals. The mean group size varied from 9.1 individuals (SD=1.6) during the two mating seasons to 12.7 individuals (SD=1.8) following the 1992 birth season. On average, the groups contained 3.2 adult males and females (SD=0.8) at weaning, 3.4 adult females (SD=0.5) and 3.6 (SD=0.9) adult males at mating, and 3.7 adult males and females (SD=0.4) following the 1992 birth season. All groups were multimale-multifemale except one that consisted of one adult pair with a juvenile and an infant. In February 1992, this small group (GP) separated from GT and established a new home range at the west corner of the site. Three months later, this small group had incorporated two adult males and two juvenile females from groups GV and GT. Throughout all phases of the reproductive cycle, mature females formed the core of their groups. In contrast, adult males moved between groups, especially during the mating season (two in 1992 and five in 1993). We also observed the exclusion of some individuals: a young female and an old male in 1992, and two old males and one young male that died from pulmonary edema in 1993. Juvenile males and females dispersed from their natal group as early as 1 year of age, during the birth season (three cases), at 112 years of age, during the mating season (five cases), or later at 2 years of age (five cases). During the 1992 and 1993 mating seasons, the operational sex ratio (i.e., potentially reproducing individuals) varied from one to 1.08 male/female. There were twice as many old males as old females (born before 1987) and only half as



Adult males Adult females Juvenile males Juvenile females Infant males Infant females Group size GT

GV

GT

29 1 1 2.5 7

2 3 1 0 2 0 8

3 4 2 0 2 2 13

3 1 0 3 1 0 8

May–July 92

40 1 2.5 1.2 4

4 4 2 1 3(2) 1 15

4 4 3 0 0 4(1) 15

3 3 0 1 3 0 10

Sept–Oct 92

GP, new group issued from GTA=GT+GP; (n), individual found dead or disapeared; ND, not determined.

30 1 1 1.67 4 GT/GP

3 3 2 1 2 1(1) 12

Adult males Adult females Juvenile males Juvenile females Infant males Infant females Group size GV

GP

Population size Adult M/F ratio Juvenile M/F ratio Infant M/F ratio Individual transfer Group fission

4 4 1 1 2 2 14

Adult males Adult females Juvenile males Juvenile females Infant males Infant females Group size GP

Population

1 1 0 1 1 0 4

Age-sex classes

Group identity

Jan–Feb 92

37 1 2.5 1 2

4 4 2 1 1 1 13

4 4 3 0 0 3 14

3 3 0 1 3 0 10

March–Apr 93

TABLE II. Group Composition and Demographic Characteristics of the Population Across Seasons and Years

35 0.91 2 1 5 GV1/GV2

3 4 2 0 1 1 11

5(1) 4 3(1) 2 0 3 8þ7

3 3 0 0 3 0 9

May–June 93

50 1.23 2.67 ND ND GP1/GP2

5 4 3 1 2 2 17

4 4 2 1 1 1 13

7 5 3 1 2 2 11 þ 9

Sept–Oct 2000


Black Lemur Demography and Behavior / 305 TABLE III. Mean Weights (in g) of Males and Females From Infancy to Adulthoodn

Age (in months) Male Female

4

8

933756 (3) 13117141 (7) 950750 (2) 13637138 (6)

16

20

18007100 (3) 16927144 (3)

32

2017778 (3) 2283722 (3) 20677122 (3) 2283744 (3)

n

Figures in parenthesis indicate the number of individuals sampled.

many young males as young females (born in 1987 or after). In 1992, mating occurred from 24 May until 3 August. In 1993, mating started as early as 7 April and lasted until 23 June. Thus, the mating season may span 4 months, from early April to early August. All females gave birth in 1991 and 1992. In 1992, eight females (73%) gave birth within a 3-week period (4–26 September), and three young females were still pregnant on 6 October. Considering that gestation lasts 4 months for this species, this means that conception occurred mainly in May and was probably delayed for the juveniles whose first reproductive activity and first parturition occurred at ~20 months and 2 years of age, respectively. The infant sex ratio at birth showed a male bias (1.67 in 1991, and 1.2 in 1992). A 4-month-old infant died in February 1992, and three infants less than 6 months old disappeared after the 1992 birth season. Thus the female reproductive rate was high (100%), but infant mortality ranged from 12.5% to 27.3%. Despite the small sample of infants born in 1991, survival into adulthood was estimated at about 75%. Between 1992 and 1993, the juvenile sex ratio varied from one to two males/female. No sex differences in the weight of young individuals could be detected (Table III). The mean adult weights of males (n=8) and females (n=8) in the middle of the 1992 wet season were 2,197 g (SD=98; range=1,975–2,400 g) and 2,148 g (SD=116; range=1780–2350 g), respectively. Although the population had increased (to 50 individuals) by 2000, the group size had not (four groups of nine to 17 individuals, mean=12.5, SD=2.5). There was an imbalance in adult and juvenile sex ratios (1.23 and 2.66 males/female, respectively). The sex ratio at birth was not determined, because not all females (77%) gave birth before the end of the study. Overall, between 1993 and 2000 the survival rate was significantly greater for males (64%) than for females (33%). In 2000, at least two 10-year-old males and two 12-year-old females were still alive. By that time, five males and two females that were captured as adults in 1987 had disappeared, which suggests that longevity for this species is about 15 years. The population comprised three to four social groups after the splitting of group GP. All observed cases of group fission occurred when group size exceeded 16 individuals (i.e., GT before separation in 1992, GV before the deaths of two males in 1993, and GP before the birth of four infants in 2000). If we compare female core areas between 1992–1993 and 2000, two adult females were still in their original home range, whereas one female had transferred from GT to GP, as did a female in the past. These latter cases indicate that during their lifespan, adult females are able to join another group or establish a new home range.

Seasonal Variation in Ranging Patterns, Activity Budgets, and Social Behavior Data accumulated from the three groups in different seasons (92 hr) in 1992 indicated that the population ranged over 33.5 ha, and the mean home range area was 18.2 ha (range=14.4–23.8 ha; convex polygon excluding sea areas; Fig. 2). The core areas of GV (24 ha), GT (16 ha), and GP (14 ha) were in the south, north, and


306 / Bayart and Simmen N

N

Sea

N

Sea

Sea GT

Mangroves

GT

Mangroves

GP

GT

Mangroves

GP

GP

Sea

100m

GV

GV

1992 all seasons

Sea

100m

1993 mating season

Sea

100m

GV

2000 birth season

Fig. 2. Group home ranges at different periods.

west, respectively. With the exception of the mangroves used only by group GT, home ranges showed considerable overlap between groups: 18% of the total area was shared by the three groups (GT and GV shared 21%, GT and GP shared 25%, and GP and GV shared up to 52%). The mean home range size was smaller during the 1993 mating season (85 hr): 11.3 ha (range=7.3–16.2 ha), and during the 2000 birth season (56 hr): 9 ha (range=5.1–13.1 ha). In 1993, the total population range was only 20 ha, but 25% of the area was still shared by the three groups. In 2000, of a total range of 24 ha, only 8% of the area was common to the three groups, suggesting group avoidance during the birth season. The proportions of traveling (32%) and resting (30–33%) during the daytime were fairly stable across seasons, but day range and speed of travel were reduced during the dry season (from 1,000 m at 20 m/min at weaning, to 500 m at 4 m/min at mating, down to 250 m at 2 m/min at birth). At mating and birth, there were also more intragroup interactions (14% and 19%, respectively) and less feeding behavior (16% and 15%, respectively) than at weaning (4% social interactions and 23% feeding behavior; Pearson w2=35.94, df=4, P=0.001). At birth, during two nights of full moon, we found more resting (55%) and intergroup interactions (10%) and less traveling (20%), feeding (10%), and intragroup interactions (5%) than during the day (Pearson w2=75.8, df=8, P=0.001). Intergroup interactions diminished from the rainy season (8%) to the dry season (4%), but their patterns varied according to the reproductive cycle (w2=18.84, df=3, P=0.001; Fig. 3). At weaning, the intergroup interactions were long-distance (53%) or pacific (47%) with groups foraging together on fruits of Dypsis spp. and Grewia spp. (both of which are nonlimiting resources during the rainy season). At mating, the animals engaged in intergroup displays (47%) or agonistic chases (11%) rather than longdistance calls (36%). These encounters were not associated with preferred food, such as the fruits of Coffea sp., Carica papaya, Dypsis spp., or Anacardium occidentale. At birth, intergroup agonistic interactions occurred at short distances (55%, of which 38% involved chases over patches of preferred plant species (fruits of Sorindeia madagascariensis and Albizia saman) at home-range boundaries). Based on group-chasing directions, GV (five chases) was dominant over GP and GT (two chases each) in 1992–1993, whereas GP (eight chases) was dominant over GV and GT (three chases each) in 2000. Intragroup interactions also showed significant differences between the mating and birth seasons (Pearson w2=12.67, df=3, P=0.005; Fig. 4). Male–male agonistic interactions occurred at mating


Black Lemur Demography and Behavior / 307 weaning 1992 0% 0%

long distance pacific encounter

47%

53%

agonistic display agonistic chase

mating 1992-93 11% 36%

long distance pacific encounter agonistic display

47%

6%

agonistic chase

birth 2000 31% 38%

long distance pacific encounter agonistic display

17%

14%

agonistic chase

Fig. 3. Distribution of intergroup interactions across the reproductive cycle (in percentage of scans).

(13%, n=209) and at birth (6%, n=108), while female–female targeted aggression was observed at birth (10%, n=108). In all seasons, females were dominant over males (displacement during foraging bouts, threats, or slapping of males, 9.5%, n=327). Male agonistic behaviors against females were never observed.

DISCUSSION

Densities, Group Sizes, and Home Ranges As with other Eulemur species, black lemur population densities vary with the level of forest disturbance [Mittermeier et al., 1994]. Overall, in 1992, densities ranged between 40 individuals/km2 in Ambalahonko ‘‘fragmented forest’’ to 200/km2 in Ambato Massif ‘‘well established secondary forest’’ [Colquhoun, 1993], with intermediate values for Ampasikely (60/km2) and Ampangorina (132/km2), two anthropogenically disturbed forests. At Ampasikely, the population declined from 100 individuals/km2 in 1987 to 88/km2 in 1988, and to 60/km2 in 1992, but rose again to 100 individuals/km2 in 2000, after the site became protected. Despite the protected status of the lemurs in northwest


308 / Bayart and Simmen weaning 1992 10%

30% 60% 0% mating 1992-93 2% 8% 13%

77% birth 2000 10% 10% 6%

74% Fig. 4. Distribution of intragroup interactions across the reproductive cycle (in percentage of scans). Pacific=affiliative interaction between two individuals; other categories=agonistic interaction between two individuals (M4M (male against male), F4M (female against male), and F4F (female against female)).

Madagascar [Harpet et al., 2000], these results indicate how vulnerable the species is in areas where deforestation and occasional hunting persist. The distribution of food resources is of paramount importance in terms of foraging and ranging behavior, and consequently the group size of primates [Oates, 1987]. In black lemurs, group size and home range appear to vary depending on the floral composition of the forest, the abundance of cultivated and introduced plant species, and provisioning. Overall, in 1992, the average group sizes were 7.4 individuals (n=11; SD=1.4) at Ambalahonko (before birth), 7.7 (n=53) in Lokobe primary forest [Andrews & Birkinshaw, 1998), 10.25 (n=4; SD=1.25) at Ambato Massif (after birth [Colquhoun, 1993]), 12.7 (n=3; SD=1.8) at Ampasikely (after birth), and 22 (n=3; SD=7.3) at Ampangorina (before birth). The protection and feeding of the lemurs at Ampangorina undoubtedly account for the largest group size observed. The floral richness of the forest at Ambato [Colquhoun, 1998a], and the year-round availability of large clumps of fruits at



Ampasikely may be responsible for the larger group sizes in these two sites compared to those observed in the Lokobe primary forest or the Ambalahonko fragmented forest. Groups may be smaller in Lokobe because food resources are more randomly distributed (Andrews, unpublished results), and in Ambalahonko because of food scarcity. Also probably linked to food patches, the composition of black lemur study populations varied between two to five groups (two to four this study; three to four at Ampangorina and in the metapopulation of 11 groups at Ambalahonko, and four to five at Ambato). Group fissions occurred when the groups reached a critical size (n=16–17, this study; n=12 in Colquhoun [1993]). At Ampasikely, the patches of cultivated plants may also be responsible for the large home range size: 18.2 ha/group (range=14.4–23.8 ha) compared to 5.25 ha/ group (range=3.5–7 ha) at Ambato (n=4 [Colquhoun, 1993]), and just 3.25 ha/ group at Lokobe (Birkinshaw, personal communication). In our study, home range size decreased during the dry season, but even with fewer hours of observation, the home ranges were still larger than in primary or well-established secondary forests.

Group Composition and Sex Ratio Regulation The groups contained on average the same number of adult females and males (respectively 2.4 and 2.3 in Andrews (unpublished results), 2.6 and 2.8 in Colquhoun [1993], and 3.4 and 3.6 this study). However, in 2000, when the Ampasikely population reached 50 individuals, we observed more adult males than adult females per group on average (4 and 3.25, respectively), and a strongly male-biased sex ratio in juveniles (2.7). To a lesser extent, in 1992 and 1993, infant and juvenile sex ratios were also male-biased. Thus, contrary to the conclusions of Andrews (unpublished results) and Colquhoun [1993], our results indicate the presence of a male-biased sex ratio in E. m. macaco, as reported earlier by Petter [1962] and Jolly [1966]. Since the latter studies (like ours) were conducted in more degraded forest habitats compared to those studied by Colquhoun and Andrews, it is possible that the type of habitat has an indirect impact on sex ratio regulation in this species. Indeed, according to Andrews (unpublished results), there is a strong infant male bias (3.5) in the five groups located on Nosy Faly Peninsula (anthropogenically disturbed forests), and a strong infant female bias (3.5) in seven groups in Manongarivo (primary forests). The local resource competition model for facultative sex ratio adjustment [Clark, 1978; Jolly, 1984; Perret, 1990; Nunn & Pereira, 2000] may explain these discrepancies. Lack of resources may have more drastic effects on maternal investment in lemurs than in other primates [Pereira et al., 1999]. In the context of secondary forests, despite potential food richness [Ganzhorn, 1995], local female–female competition may occur because of a larger group size and higher number of females per group, as observed for black lemurs. Considering that the optimal number of females per group is the result of an evolutionary adjustment of group size to food availability in undisturbed forests, a higher number of females would enhance local female–female competition. This would lead to higher mortality rates in females, as detected in the present study. Resource competition between females would be lowered with a high production of males. Consequently, if the average number of females per group is below 3, as in Lokobe (2.4), Manongarivo (2.4), and Ambato (2.6), we observe a female-biased sex ratio at birth. If it is over 3, as in Ampasikely (3.4), Nosy Faly Peninsula (3.4), and Nosy Komba (6), we observe a male-biased sex ratio at birth. Moreover, this model agrees with observations made in captive lemur populations. In zoos where



females were grouped, the sex ratio was often male-biased, whereas in a semifree-ranging group of black lemurs, targeted aggression and exclusion between females occurred whenever the number of adult females in the group rose above 3 (Roeder, personal communication). In captive brown lemurs as well, targeted aggression has been linked to group size and adult sex ratio [Vick & Pereira, 1989].

Seasonal Variation in Behavior Our data on reproductive parameters confirm both seasonal breeding and birth synchrony for this species [Rasmussen, 1985]. Moreover, the patterns of intergroup encounters varied across seasons (pacific during the rainy season, and agonistic with dominance of larger groups over smaller groups during the dry season). Dry-season observations indicate that the black lemurs did not suffer from food scarcity, because numerous fruiting or flowering cultivated species were available (Simmen et al., unpublished results). At mating, intergroup agonistic encounters were not linked to the principal food, and probably reflect male–male competition for mates, whereas at birth, disputes over preferred fruit species indicate possible female–female competition for resources during lactation. In the future, social competition must be studied more extensively along with the distribution and productivity of food resources. In our study, black lemurs behaved during the dry season as if they were minimizing foraging costs in order to allocate more energy for reproductive purposes [Pereira et al., 1999]. However, more data are needed regarding nocturnal activity [Andrews & Birkinshaw, 1998; Colquhoun, 1998b] and reproductive tactics in this species to confirm this hypothesis. ACKNOWLEDGMENTS We are grateful to the Ministe`re des Eaux et Foreˆts in Madagascar, and to R. and H. D’Ambelle for permission to conduct this research. We thank Y. Rumpler and B. Krafft for supporting the project. We appreciate the help and advice of J.J. Petter, R.D. Martin, B. Meier, S. Crovella, D. Montagnon, B. Rakotosamimanana, C. Rabarivola, the technicians from Tsimbazaza (Antananarivo) and CNRO (Nosy Be), the students, the forest guides, and the villagers. We thank the anonymous reviewers who commented on earlier drafts of the manuscript.

REFERENCES Altmann J. 1974. Observational study of behavior: sampling methods. Behaviour 49: 227–267. Andrews JR, Birkinshaw CR. 1998. A comparison between the daytime and nighttime diet, activity and feeding height of the black lemur, Eulemur macaco (Primates, Lemuridae), in Lokobe Forest, Madagascar. Folia Primatol 69(suppl 1):175–182. Bayart F, Rabarivola C, Ludes B, Krafft B, Rumpler Y. 1993. Eco-ethological, demographic and genetic survey of three populations of Eulemur macaco macaco in northwestern Madagascar. In: Proceedings of the 23rd International Ethological Conference, Torremolinos, Spain. 49p.


Birkinshaw CR. 1999. The importance of the black lemur (Eulemur macaco) for seed dispersal in Lokobe Forest, Nosy Be. In: Rakotosamimanana B, Rasamimanana H, Ganzhorn JU, Goodman SM, editors. New directions in lemur studies. New York: Kluwer Academic/Plenum Publishers. p 189–199. Clark AB. 1978. Sex ratio and local competition resource in a prosimian primate. Science 201:165–168. Colquhoun IC. 1993. The socioecology of Eulemur macaco: a preliminary report. In: Kappeler PM, Ganzhorn JU, editors. Lemur social systems and their ecological basis. New York: Plenum Press. p 11–23.

Black Lemur Demography and Behavior / 311 Colquhoun IC. 1998a. The lemur community of Ambato Massif: an example of the species richness of Madagascar’s classified forests. Lemur News 3:11–14. Colquhoun IC. 1998b. Cathemeral behavior of Eulemur macaco macaco at Ambato Massif, Madagascar. Folia Primatol 69(suppl 1):22–34. Crook JH, Aldrich-Blake P. 1968. Ecological and behavioral contrasts between sympatric ground dwelling primates in Ethiopia. Folia Primatol 8:192–227. Ganzhorn JU. 1995. Low-level forest disturbance effects on primary production, leaf chemistry, and lemur populations. Ecology 76:2048–2096. Gould L, Sussman RW, Sauther ML. 2003. Demographic and life-history patterns in a population of ring-tailed lemurs (Lemur catta) at Beza Mahafaly Reserve, Madagascar: a 15-year perspective. Am J Phys Anthropol 120:182–194. Harpet C, Jeannoda V, Hladik CM. 2000. Sites ` le´muriens sacre´s en pays Sakalava, au a nord-ouest de Madagascar: reáctualisation des donneés et implications dans les programmes de de´veloppement et de conservation. Rev Ecol (Terre Vie) 55:291–295. Jolly A. 1966. Lemur behavior: a Madagascar field study. Chicago: University of Chicago Press. 187p. Jolly A. 1984. The puzzle of female feeding priority. In: Small MF, editor. Female primates: studies by women primatologists. Monographs in primatology. Vol. IV. New York: Alan R. Liss. p 197–215. Jolly A. 1998. Pair bonding, female aggression and the evolution of lemur societies. Folia Primatol 69(suppl 1):1–13. Jolly A, Dobson A, Rasamimanana HM, Walker J, O’Connor S, Solberg M, Perel V. 2002. Demography of Lemur catta at Berenty Reserve, Madagascar: effects of troop size, habitat and rainfall. Int J Primatol 23:327–353. Kappeler PM, Ganzhorn JU. 1994. The evolution of primate communities and societies in Madagascar. Evol Anthropol 2:159–171. Kappeler PM. 1997. Determinants of primate social organization: comparative evidence and new insights from Malagasy lemurs. Biol Rev 72:111–151. Kappeler PM. 1999. Lemur social structure and convergence in primate socioecology. In: Lee PC, editor. Comparative primate socioecology. Cambridge: Cambridge University Press. p 273–299. Kappeler PM. 2000. Causes and consequences of unusual sex ratios among lemurs. In: Kappeler PM, editor. Primate males. Cambridge: Cambridge University Press. p 55–63.

Meier B, Rumpler Y. 1992. Enzyme variability in island and mainland populations of black lemur (Eulemur macaco): a contribution to conservation biology. Karger Gazette 54:10–12. Mittermeier RA, Tattersall I, Konstant WR, Meyers DM, Mast RB. 1994. Lemurs of Madagascar. Conservation international tropical field guide series 1. Washington, DC: Conservation International. 356p. Nunn CL, Pereira ME. 2000. Group histories and offspring sex ratios in ringtailed lemurs (Lemur catta). Behav Ecol Sociobiol 48: 18–28. Oates JF. 1987. Food distribution and foraging behaviour. In: Smuts BB, Cheney DL, Seyfarth RM, Wrangham R, Struhsaker TT, editors. Primate societies. Chicago: Chicago University Press. p 197–209. Overdorff DJ, Merenlender AM, Talata P, Telo A, Forward ZA. 1999. Life history of Eulemur fulvus rufus from 1988–1998 in southeastern Madagascar. Am J Phys Anthropol 108:295–310. Overdorff DJ, Johnson S. 2003. Eulemur, true lemurs. In: Goodman SM, Benstead JP, editors. The natural history of Madagascar. Chicago: University of Chicago Press. p 1320–1324. Pereira ME. 1993. Seasonal adjustment of growth rate and adult body weight in ringtailed lemurs. In: Kappeler PM, Ganzhorn JU, editors. Lemur social systems and their ecological basis. New York: Plenum Press. p 205–221. Pereira ME, Strohecker RA, Cavigelli SA, Hughes CL, Pearson DD. 1999. Metabolic strategy and social behavior in Lemuridae. In: Rakotosamimanana B, Rasamimanana H, Ganzhorn JU, Goodman SM, editors. New directions in lemur studies. New York: Kluwer Academic/Plenum Publishers. p 93–117. Perret M. 1990. Influence of social factors on sex ratio at birth, maternal investment and young survival in a prosimian primate. Behav Ecol Sociobiol 27:447–454. Petter JJ. 1962. Recherches sur l’ećologie et l’e´thologie des le´muriens malgaches. Me´m Mus Natl Hist Nat Paris Se´r A Zool XXVII:1–146. Rabarivola C, Meier B, Langer C, Bayart F, Ludes B, Rumpler Y. 1998. Comparison of genetic variability in wild insular and mainland populations of Eulemur macaco: implications for conservation strategy. Folia Primatol 69(suppl 1):136–146. Rasmussen DT. 1985. A comparative study of breeding seasonality and litter size in eleven taxa of captive lemurs (Lemur and Varecia). Int J Primatol 6:501–517. Simmen B, Tarnaud L, Bayart F, Hladik A, Thiberge A-L, Jaspart S, Jeanson M, Marez


312 / Bayart and Simmen A. Richesse en me´tabolites secondaires des foreˆts de Mayotte et de Madagascar et incidence sur la consommation de feuillage chez deux espe`ces de le´murs (Eulemur spp.). Rev Ecol (Terre Vie) (in press). Sussman RW. 2002. Adaptive array of lemurs of Madagascar revisited. Evol Anthropol Issues News Rev 11(Suppl 1):75–78. Tattersall I. 1977. Ecology and behavior of Lemur fulvus mayottensis (Primates, Lemuriformes). Anthropol Pap Am Mus Nat Hist 54:422–482. van Schaik CP, Kappeler PM. 1993. Life history, activity period and lemur social systems. In: Kappeler PM, Ganzhorn JU, editors. Lemur social systems and their ecological basis. New York: Plenum Press. p 241–260.


van Schaik CP, Kappeler PM. 1996. The social systems of gregarious lemurs: lack of convergence with anthropoids due to evolutionary disequilibrium? Ethology 102: 915–941. Vasey N. 2003. Varecia, ruffed lemurs. In: Goodman SM, Benstead JP, editors. The natural history of Madagascar. Chicago: University of Chicago Press. p 1332–1336. Vick LG, Pereira ME. 1989. Episodic targeting aggression and the histories of Lemur social groups. Behav Ecol Sociobiol 24: 265–276. Wright PC. 1999. Lemur traits and Madagascar ecology: coping with an island environment. Yearb Phys Anthropol 42: 31–72.