Introduction

Since 1989, we have been conducting a socio-ecological study on wild ring-tailed lemurs (Lemur catta), based on individual identification, at the Berenty Reserve, Madagascar (Koyama et al. 2002). Ring-tailed lemurs form female-bonded/matrilineal social groups, and typically participate in two types of competition for resources, i.e., between-group competition and within-group competition (Koyama et al. 2005; Takahata et al. 2005).

It is well known that there are negligible sexual differences in the body mass of prosimian species (Kappeler 1990). However, there are few data on the body mass of wild and individually identified populations of lemurid species (Sussman 1991; Richard et al. 2000). In November 1999, we captured 101 ring-tailed lemurs inhabiting our study area (14.2 ha), in order to collect blood samples for analysis of their genetic variability. We also measured their body mass, and counted the ticks [Haemaphysalis (Rhipistoma) lemuris Hoogstraal, 1953; Uilenberg et al. 1979] infesting their facial skin and external auditory meatuses. In this report, we describe a preliminary analysis of sexual dimorphism, correlations of rank with body mass and tick infestation, and differences between groups in body mass and tick infestation.

Methods

Since 1989, a wild population of ring-tailed lemurs has been individually identified at the Berenty Reserve, Madagascar (Koyama et al. 2002, 2005). In November 1999, from the 98 individuals of Troops C1, C2A, C2B, CX, T1, and T2, 95 lemurs were anesthetized with ketamine hydrochloride by using blowpipes and darts or a trap. A further six adult males were captured; four of them had formerly belonged to one of our study groups, and the other two had followed one of our study groups. We measured their body mass, and counted the number of ticks infesting their facial skin and external auditory meatuses. The tick has been identified as [Haemaphysalis (Rhipistoma) lemuris Hoogstraal, 1953; Uilenberg et al. 1979; Takahata et al. 1998]. At Berenty, there is also a large black tick, but this was not included in our study.

The captured individuals comprised 29 adult females (3 or more years of age), 7 subadult females (2 years of age), 6 juvenile females (1 year of age), 10 infant females, 31 adult males (3 or more years of age), 8 subadult males (2 years of age), 5 juvenile males (1 year of age), and 5 infant males. At Berenty, births occur between late August and late December (Koyama et al. 2001). In November 1999, all births had occurred before the date of capture.

Of the captured lemurs, 75 individuals were identified from the year of birth, and therefore their ages were known. Seven adult females have been observed since 1989 and were confirmed to have attained the age of 13 years or more in 1999. Two adult females and 17 adult males had immigrated into the study groups from unknown groups, and their ages were all unknown.

In each study group, dominance ranks among the adult members were determined on the basis of (1) approach–retreat interactions while feeding and drinking, and (2) submissive vocalization (Koyama et al. 2005). In every group, adult females were dominant over adult males. The relative ranks of adult females and males were calculated using the following formula, which indexed the alpha animal as 100% and the lowest ranking animal as 0%.

$$ {\text{Relative}}\;{\text{rank}}\;{\text{of}}\;{\text{an}}\;{\text{adult}}\;{\text{female/male}}\;{\text{(\%)}}\;{\text{ = }}\;{\text{(}}N\; - \;R{\text{)}}/(N\; - \;1)\;{\text{ $ \times $ }}\;{\text{100}} $$
N :

number of adult females/males in the group

R :

rank of an individual among same-sex members

In each group, females/males were grouped into four rank categories: alpha (relative rank = 100%), high-ranked (99.9–66.7%), mid-ranked (66.6–33.3%), and low-ranked individuals (<33.3%). Statistical analyses were performed using Excel 2002 (Microsoft 2001) and Statistica (StatSoft Inc., 1999). The level of significance was P < 0.05, and all tests were two-tailed.

Results

Body mass and age

Figure 1 shows the age (years) and body mass (kg) for the 75 individuals of known age. Note that our data were not of a longitudinal type. Up to the age of 3 years, there were significant correlations between age and body mass (females, n = 27, r = 0.963, P < 0.001; males, n = 23, r = 0.897, P < 0.001). Thereafter, body mass appeared to reach a growth plateau. No significant correlation between age and body mass was observed in 4- to 10-year-old adult females (n = 16, r = 0.479, P > 0.05). For adult males of between 4 and 12 years of age, there was a significant correlation in this regard (n = 9, r = 0.685, P < 0.05), but the rate of increase was very small.

Fig. 1
figure 1

Age (years) and body mass (kg) for the 75 individuals of known ages, with the least square smoothing curves. The body mass of 15 adult females and 17 adult males of unknown age is also represented

Full size image

Including adult individuals of unknown ages, there were no sexual differences in the body mass of adults (29 females and 31 males; Mann–Whitney U test, U = 354, z = 1.134, P > 0.1). The mean body mass of adult females was 2.27 kg (n = 29, SD = 0.20) and that of adult males was 2.22 kg (n = 31, SD = 0.22); these values are almost equal to the body masses of 2.207 kg (adult females, n = 24) and 2.213 kg (adult males, n = 41) recorded in the wild population of Beza Mahafaly Reserve, Madagascar (Sussman 1991), but smaller than the 2.68 kg and 2.71 kg recorded in the captive population of Duke University Primate Center (Kappeler 1991).

There were no sexual differences in the body mass of subadults (2 years of age) (7 females and 8 males, U = 16.5, P > 0.1) and juveniles (1 year of age) (6 females and 5 males, U = 7.5, P > 0.1). The mean body mass was 1.9 kg (SD = 0.09) for subadult females, 1.84 kg (SD = 0.32) for subadult males, 1.3 kg (SD = 0.14) for juvenile females, and 1.42 kg (SD = 0.08) for juvenile males.

On the other hand, there was a sexual difference in the body mass of infants. Infant females were significantly heavier than infant males (10 females and 5 males, U = 7.5, P < 0.04). Figure 2 shows the correlations between age (days) and body mass (g) of female and male infants. The body mass of infant females seemed to be linearly correlated with days after birth (n = 10, Pearson’s r = 0.894, P < 0.01), but the sample size for infant males was too small to analyze (n = 5, r = 0.660, P > 0.2).

Fig. 2
figure 2

Age (days) and body mass (g) of infants (females, y = −33.99 + 6.6x; males, y = 122.4 + 2.82x)

Full size image

Body mass and rank of adults

Overall, there was no significant difference in the body mass of adult lemurs among the five study groups [one-way ANOVA test, F (5, 48) = 1.28, P > 0.2] (Fig. 3). The post hoc analysis exhibited that there was a significant difference in body mass of adults between Troop CX and Troop C2B (Duncan’s multiple range test, P < 0.03). There was no significant difference in other pairs of groups.

Fig. 3
figure 3

Mean body mass of adult members of each study group [one-way ANOVA, F (5, 48) = 1.28, P > 0.2]

Full size image

Based on the pooled data of all the study groups, there was no significant difference in body mass among female rank categories [F (3, 25) = 1.60, P > 0.2]. Figure 4a shows that there was no consistent correlation between female rank and body mass. The post hoc analysis exhibited that there was a significant difference in body mass between alpha females and high-ranked females (Duncan’s multiple range test, P < 0.05). On the other hand, there was no significant difference among alpha, middle-ranked, and low-ranked females.

Fig. 4a, b
figure 4

Body mass and rank categories. a Adult females [F (3, 25) = 1.60, P > 0.2], b adult males [F (3, 20) = 4.02, P < 0.03]

Full size image

There was a significant difference in the body mass among male rank categories [F (3, 20) = 4.02, P < 0.03] (Fig. 4b). In particular, the low-ranked males exhibited the lowest values of body mass, and were significantly lower than those of alpha males (Duncan’s multiple range test, P < 0.001).

From 1998 to 1999, nine males migrated into the study groups. The mean body mass of these newcomers was slightly smaller than that of the resident males (2.14 kg vs 2.27 kg), but the difference was not significant (9 newcomers and 16 resident males, U = 43, z = 1.654, P > 0.05).

Is the body mass of immature individuals affected by their mother’s rank? Owing to the small sample size, in order to answer this question, the data for immature males and females were pooled to facilitate calculation (Fig. 5). There were no significant differences in the body mass of subadults among their mothers’ ranks [H (3, 15) = 3.116, P > 0.3]. Similarly, there were no significant differences in the body masses of juveniles [H (3, 11) = 2.434, P > 0.4] and those of infants [H (3, 15) = 1.839, P > 0.6] among their mothers’ ranks.

Fig. 5
figure 5

Body mass of juveniles and their mothers’ ranks [Kruskal–Wallis test, H (3, 15) = 3.116, P > 0.3 (2 years); H (3, 11) = 2.434, P > 0.4 (1 year); H (3, 15) = 1.839, P > 0.6 (infant)]

Full size image

Number of ticks, age, rank, and body mass

There was a significant difference in the number of ticks found in adult lemurs among the six study groups [H (5, 54) = 22.861, P < 0.001]. Figure 6 shows that Trop CX was severely infested with ticks. Similar group differences were found for immature members of six study groups [H (5, 41) = 15.493, P < 0.01].

Fig. 6
figure 6

Mean number of ticks infesting the facial skin and external auditory meatuses of adult members of each study group [H (5, 54) = 22.861, P < 0.001]

Full size image

On the other hand, there was still a significant difference in the number of ticks found in adult lemurs among Troops C1, C2A, C2B, T1, and T2 [H (4, 49) = 12.794, P < 0.03]. Figure 6 shows that the adult lemurs of Troops C1, T1, and T2 were moderately infested with ticks, and that those of Troops C2A and C2B exhibited a lower tick infestation. Thus, the study Troops could be divided into three groups with respect to tick infestation; (1) Troop CX, (2) C1, T1, and T2, and (3) C2A and C2B.

The lemurs of Troop CX were infested with 30.8 ticks on average (n = 11; range 5–75 ticks). There was no significant difference in infestation between the age–sex classes within this group [H (4, 11) = 3.212, P > 0.5] (Fig. 7a).

Fig. 7a–c
figure 7

Mean number of ticks found on the skin and external auditory meatuses of different age-sex classes. a Troop CX [Kruskal–Wallis test, H (4, 11) = 1.984, P > 0.7), b Troop C1, T1, and T2 (H (6, 61) = 20.81, P < 0.01), c Troop C2A and C2B (H (6, 23) = 9.62, P > 0.1]

Full size image

The lemurs of Troops C1, T1, and T2 were infested with 4.3 ticks on average (n = 61; range 0–31). There were significant differences in the number of infesting ticks among the age–sex classes. Figure 7b shows that adult males and subadults of both sexes were infested with more ticks than infants, juveniles, and adult females [H (6, 61) = 20.81, P < 0.01].

In Troops C1, T1, and T2, there was no significant correlation between body mass and the number of infesting ticks for adult females (n = 17, r = 0.3847, P > 0.1). In contrast, there was a significant correlation between body mass and the number of infesting ticks for adult males (n = 16, r = 0.7594, P < 0.01). There was no significant difference between the newcomers and resident males in the number of ticks (n 1 = 5, n 2 = 11; U = 24, P > 0.6).

In Troops C1, T1, and T2, there was also no significant difference in the number of infesting ticks among the rank categories of adult females [one-way ANOVA test, F (3, 13) = 0.26, P > 0.8] or among the rank groups of adult males [F (3, 11) = 0.68, P > 0.5] (Fig. 8). As shown in Fig. 8b, there was a great variation in the number of ticks among higher-ranking males; however, the reason for this is uncertain.

Fig. 8
figure 8

Mean number of ticks per adult rank categories in Troops C1, T1, and T2. a Females [F (3, 13) = 0.26, P > 0.8]; b males [F (3, 11) = 0.68, P > 0.5]

Full size image

The lemurs of Troops C2A and C2B were infested with 1.0 ticks on average (n = 23; range 0–6). There were no significant differences in the mean number of infesting ticks among age–sex classes for them [H (6, 23) = 9.62, P > 0.1] (Fig. 7c).

Table 1 shows the reproductive parameters of the study groups recorded from 1997 to 2000. Apparently, the females of Troop C2A, who were the heaviest and were infrequently infested with ticks, did not always attain high reproductive success.

Table 1 Body mass and number of ticks of adult members of each study group, and reproductive parameters recorded from 1997 to 2000
Full size table

Discussion

Body mass of ring-tailed lemurs

In most anthropoid primates, adult males are larger than adult females. This type of sexual dimorphism has been discussed from the viewpoints of ecology, physiology, and life history (Masterson and Hartwig 1998; Plavcan 2001). In particular, it has been disputed whether such a dimorphism is a natural consequence of sexual selection (i.e., male–male competition for access to mates) (Plavcan and van Schaik 1997).

In contrast, there are few sexual differences in the body masses of prosimian species, regardless of their social systems (Kappeler 1990, 1991; Terranova and Coffman 1997; Wright 1999; Müller 1999; Sauther et al. 2002; Cuozzo and Sauther 2004). In the present study, we found no sexual difference in the body mass of adult ring-tailed lemurs; this is consistent with observations from previous studies (e.g., Kappeler 1991). With the exception of infants, both males and females showed similar growth curves, although our data were not of a longitudinal type. The body mass of the lemurs appears to increase up to the age of 3 years in Berenty, this is the age when most females enter the reproductive community and most males leave their natal groups (Koyama et al. 2001, 2002). At 3 years of age, the body mass of both males and females has probably reached a plateau. Figure 1 suggests that male ring-tailed lemurs neither show extended growth (bimaturism) nor grow faster than females (rate dimorphism). This is different from the growth pattern observed in males of anthropoid species (Leigh 1992, 1995; Turner et al. 1997; Hamada et al. 2004).

The present data show that there is no consistent correlation between the body mass of adult females and their ranks. In contrast, higher-ranked males tended to be heavier than lower-ranked males. However, we cannot conclude that higher-rank originates from body mass and vice versa, because of the lack of continuous data. In severe aggressive interactions between males, heavier males should have an advantage over others in male–male competition for female mates (e.g., Koyama 1988). However, another question may arise. It should be puzzling why male ring-tailed lemurs, despite their polygynous mating system, do not develop sexual dimorphism, since an individual’s ability to compete should be directly related to the outcome of the fight in a multi-male and multi-female social structure (Wright 1999; Koyama et al. 2005). Unfortunately, at present we have insufficient data to make a thorough investigation of this question.

Ring-tailed males frequently transfer between groups. At Beza Mahafaly, Sussman (1991) pointed out that males in the process of migrating might be under nutritional stress. Then one might suppose the newcomers to be smaller than the resident males. In our data, however, there was no significant difference between the newcomers and the resident males.

Infection by ticks

Many species of ectoparasites (e.g., lice and ticks) infest the bodies of primates (e.g., Rajagopalan and Anderson 1971, for Presbytis entellus and Macaca radiata; Rijksen 1978, for orangutans; Goodall 1986, for chimpanzees). Evolutionary biologists have long been interested in the coevolution of hosts and parasites (e.g., Futuyma 1986). In particular, behavioral ecologists have discussed the hypothesis that parasites play a crucial role in sexual selection (e.g., Zuk et al. 1990). On the other hand, primatologists have pointed out the hygienic function of grooming, that is, the removal of ectoparasites (Tanaka and Takefushi 1993, for Japanese macaques). Recently, Nunn and his colleagues have analyzed a data set of host characteristics and parasite diversity in primates (Nunn et al. 2003; Vitone et al. 2004). They pointed out that several key features of host biology (body mass, diet, sociality and ranging behavior) accounted for most variation in parasite species richness.

In Berenty, a tick species (H. lemuris) was present on the naked skin of L. catta (Takahata et al. 1998). The same tick species was found to infest the bodies of ring-tailed lemurs at Beza Mahafaly (Sauther et al. 2002), and those of ruffed lemurs (Varecia variegata and V. rubra) at Masoala Peninsula, Betampona Strict Nature Reserve, and Mantadia National Park (Junge and Louis 2005).

At Beza Mahafaly, Sauther et al. (2002) found no marked differences among age classes of L. catta for the frequency of individuals with ectoparasites; however, there was a sex difference, with males having more ectoparasites than females. The present data showed several different tendencies. In the most heavily infested group (CX, Fig. 7a), and the least heavily infested groups (C2A and C2B, Fig. 7c), the age–sex classes did not differ in tick infestation level. In contrast, there was a significant difference in the number of ticks found on the facial skin among age–sex classes in three groups moderately infested with ticks (Fig. 7b); in these groups, adult males and subadults of both sexes were infested with more ticks than infants, juveniles, and adult females. This tendency may be related to the frequency of grooming within the group. Oda (1996) and Nakamichi and Koyama (1997) analyzed the social interactions of ring-tailed lemurs, and they found that adult males, in particular low-ranked individuals, were infrequently involved in grooming interactions with other adult members. In contrast, grooming occurred more frequently between closely related adult females and their offspring than between unrelated individuals. Further, subordinates were likely to groom dominants more frequently than vice versa. Thus, in moderately infested groups, it may be expected that ticks infest adult females less frequently than adult males.

Group variation

Among the study groups, there was a small difference in mean body mass, and a large difference in mean number of infection by ticks per individual (see Figs. 3, 6). The mean body mass of adult members of the smallest-bodied group (CX) was significantly lower than that of members of the heaviest group (C2A), and Troop CX members were heavily infested by ticks. Smaller body mass and heavier infestation recorded in Troop CX may be caused by their habitat (Troop CX inhabited the most humid area along the gallery forest) or their inferiority in between-group competition (Troop CX had split from Troop C1 during the period from 1993 to 1994, and they have been subordinate to neighboring groups) (Koyama et al. 2002).

Is such a group variation in body mass and tick infection correlated with fertility? For example, in Verreaux’s sifaka (Propithecus verreauxi) at Beza Mahafaly, Richard et al. (2000) observed a close link between female body mass and fertility. They concluded that the sifaka was a “capital breeder” and that a female’s body mass (store of fat) at the outset of the mating season strongly influences the probability that she will give birth in the following birth season. However, the present data suggest that there is no consistent correlation between body mass and infection by ticks and female fertility among ring-tailed lemurs of Berenty Reserve (see Table 1). The mean number of surviving infants 1 year after birth of Troop CX was almost equal to that of Troop C2A. Thus, at present, it is uncertain whether body mass and infection by ticks affects the fertility of ring-tailed lemurs.