International Journal of Primatology

, Volume 28, Issue 1, pp 159–182 | Cite as

Bimorphism in Male Verreaux’s Sifaka in the Kirindy Forest of Madagascar

Article

Male primates in species with pronounced secondary sexual adornments can exhibit reversible or irreversible bimorphism, i.e., striking variation in the degree to which males express the adornments. Verreaux’s sifaka (Propithecus verreauxi verreauxi) use scent marking as a form of communication and exhibit sex differences in scent glands. Some males exhibit a pronounced brown staining around their sternal gland, whereas others do not. We studied morphological and behavioral characteristics of males in 6 social groups in Kirindy Forest, Madagascar, from November 2000 to March 2002 to evaluate the hypotheses that the bimorphism in male sifaka chest status represents alternative mating tactics and is a badge of status. Males are clearly divided into 2 categories: clean and stained chests, with rare, but informative, intermediate males. The chest staining probably results from the males scent marking with their sternal glands, because stained-chested males scent marked significantly more often than clean-chested males. Though sample sizes are small, chest status did not appear to depend on body size. Chest status is reversible and related to dominance rank. In each group, only 1 male, the dominant, was stained-chested, whereas all other (subordinate) males were clean-chested. These findings suggest that stained chests are visual and olfactory signals of dominance rank and that clean chests signal lack of competitive intent. Thus, this bimorphism may reflect alternative mating tactics used by males to maximize their reproductive success based upon their social environment.

KEYWORDS

alternative mating strategies Propithecus Sexual selection signal 

INTRODUCTION

Many primate males exhibit secondary sexual characteristics that function as adornments, e.g., capes in hamadryas baboons (Papio hamadryas); temporal bulges in uakari (Cacajao spp.); and cheek flanges, throat sacs, and longer hair in orangutans (Pongo pygmaeus; Dixson, 1998). However, in some species, not all sexually mature males possess the adornments. For lack of an established term, we call the phenomenon (intrasexual) bimorphism to avoid confusion with sexual dimorphism. In orangutans, male bimorphism (bimaturism or arrested development: Maggioncalda et al., 1999; Utami et al., 2002) arises because males show a variable delay in the development of the adornments, and is therefore irreversible, whereas in mandrills (Mandrillus sphinx), the development of male adornments, in particular the bright coloration of the face and rump, are reversible (Setchell and Dixson, 2001b).

Bimorphism, reversible or irreversible, may reflect a conditional strategy, with alternative mating tactics based on the males’ status in the social environment (Gross, 1996). Researchers have described conditional strategies as occurring when competitively inferior males “make the best of a bad job” (Dawkins, 1980) or “make the best of a bad batch” (Shuster and Wade, 2003), depending on the underlying genetic basis for the alternative mating tactics. For example, Utami Atmoko and van Hooff (2004) interpreted the bimorphism in orangutans as an expression of alternative male mating tactics. Flanged males have fully developed secondary sexual characteristics, are highly intolerant of other flanged males—but not unflanged males—and advertise their presence to females by emitting long calls. Unflanged males are also sexually active, have not yet developed secondary sexual characteristics, are tolerant toward other males, and silently search for females. The 2 male morphs develop on different timelines (Maggioncalda et al., 2000) that may be density dependent (Utami Atmoko and van Hooff, 2004; van Schaik, 2004). Because orangutan bimaturism is irreversible, the switch between reproductive tactics can occur only once in a male’s lifetime (Utami Atmoko and van Hooff, 2004).

When bimorphism is reversible, the phenotypic variation may be a badge of status. Badges typically describe variation in the coloration of plumage (Rohwer and Ewald, 1981) and function as indicators of competitive ability that can be related to aggressiveness during feather growth (McGraw et al., 2003). Reliable signals of social status, such as badges, can be useful for dominant males as they decide which males to allow onto their territory and for subordinate males as they decide which territory to join (Rohwer and Ewald, 1981). Signaling of status creates a situation in which both dominant and subordinate males benefit from inhabiting the same territory (Rohwer and Ewald, 1981). Because feathers are replaced with each molt, badges are reversible.

Primates also have reversible badges of status. For instance, some researchers have suggested that male coloration acts as a badge of status in mandrills (Setchell and Dixson 2001a,b; Setchell and Wickings, 2005). Male vervets (Cercopithecus aethiops) show bright scrotal coloration, which also appears to act as a badge of status (Gerald, 2001). These badges play an important role in both intra- and intersexual selection (Gerald, 2003). Because the different male morphs in bimorphic species can be associated with differences in dominance rank, and possibly other features such as testosterone levels, and gregariousness (Setchell and Dixson, 2001a), bimorphism in badges may represent alternative mating tactics.

Verreaux’s sifaka (Propithecus verreauxi verreauxi) are sexually monomorphic strepsirrhines with a body mass of 2.5–4 kg (Richard et al., 2000; Lewis and Kappeler, 2005). They live in small mixed-sex groups of 2–13 individuals (Richard et al., 1993) with variable adult sex ratios (Richard, 1985). The adult males in the Kirindy Forest of western Madagascar are bimorphic: some exhibit a pronounced brown greasiness around the sternal gland and on their chests, whereas others have clean, white chests. Dominant males have higher fecal testosterone than that of subordinate males (Brockman et al., 1998, 2001; Kraus et al., 1999), which may be a result of suppression by the dominant males (Kraus et al., 1999). Sifaka have a limited breeding season; a female’s behavioral estrus ranges from .5 to 96 h (Brockman, 1999). Like many mammals, sifaka use scent-marking as a form of communication (Kraus et al., 1999; Lewis, 2005, 2006). Scent glands are sexually dimorphic: females have scent glands in the anogenital region, whereas males have both anogenital and sternal glands (Schilling, 1979).

Petter (1962) briefly described the patches of stained chest hair on museum specimens in different Propithecus. However, no one had studied this staining in detail. We document striking phenotypic variation in Verreaux’s sifaka male chest status and evaluate the hypotheses that chest status represents alternative mating tactics and that the staining of the chest is a badge of status. We predicted that (1) male social and mating behavior varies with chest status, (2) chest status is reversible, and because body size influences fighting ability (3) chest status is related to body size. We present results on dominance, mating behavior, and morphology suggesting that the bimorphism in Verreaux’s sifaka is a reversible badge of status and an expression of alternative reproductive tactics. The findings must be considered preliminary until additional hormonal and genetic data are available.
Table I.

Adult males captured in 2002 breeding season

Group

Males:Females

ID

Age

Chest status

A

2:1

LZ

>10 yr

Stained

A

2:1

BG

 9

Clean

B

1:2

BN

>10

Stained

E

3:2

DW

>7

Stained

E

3:2

DP

>7

Clean

E

3:2

AN

 6/7

Clean

H

1:1

SP

>7

Stained

J

3:3

PY

 9

Stained

J

3:3

JZ

>7

Clean

J

3:3

TK

>7

Intermediate

T

1:1

BO

>7

Stained

Solitary

n/a

KB

>5

Stained

Solitary

n/a

TT

>5

Intermediate

METHODS

Field Site

Kirindy Forest is a primary deciduous forest located at ca. 44°39′E, 20°03′S. Mean annual rainfall is ca. 800 mm. The 60-ha study area is within a forestry concession of the Centre Formation Professionelle Forestière de Morondava and part of the field station of the German Primate Centre (DPZ). Ganzhorn and Sorg (1996) and Ganzhorn (2002) provided a detailed description of the forest.

Study Population

Over the past decade, researchers have captured more than 120 sifaka at Kirindy Forest and individually marked them with radio collars (Biotrack, Wareham, Dorset, UK) or colored nylon collars. The individuals were either solitary or members of 1 of 19 social groups within the CS7 area of the forest. We conducted monthly censuses of all social groups from October 2000 to March 2002. Adults in 6 groups were the foci of behavioral observations (Table I). We captured additional individuals for identification markings and hence included them when relevant. We estimated ages of sedated individuals during our captures by evaluating tooth wear.

While Verreaux’s sifaka males may become sexually active at 3 yr (Kraus et al., 1999; Richard et al., 2002), no 3- or 4-yr-old male has ever been found to sire offspring (Lawler, 2003). Moreover, long-bone growth continues until 5 yr (Richard et al., 2000). We therefore defined adulthood in Verreaux’s sifaka for both sexes to begin at the reproductively mature age of 5 yr, per Lawler (2003). Accordingly, social groups comprised 1–3 adult females and 1–3 adult males (Table I). Subadult males (3–4-yr-olds) are immature and we did not observe them to have stained chests and thus did not include them. The mating season for sifaka at Kirindy occurs from late January through March (Kraus et al., 1999; Lewis and Kappeler, 2005). Females give birth to a single young in July or August (Lewis and Kappeler, 2005), which they carry for the first 3 mo of the infant’s life (Dill, 2000; Grieser, 1992).

Capture Methods

In early 2002, during the mating season, we captured and sedated all focal individuals and some solitary males. We sedated and measured subjects per Glander et al. (1992). We captured individuals via a blow pipe that delivers disposable nonbarbed darts with a 3/8-inch needle. We loaded darts with 0.5 ml of the injectable anesthetic GMII (Rensing, 1999). We darted individuals at distances ≤20 m, preferably injecting the hindquarters. An experienced Malagasy technician conducted all the darting. Darted individuals fell from a tree and were caught in a large cotton cloth. If the individual did not fall, an assistant climbed the tree to retrieve the subject. We darted only adults without dependent infants. We did not capture all adults at the same time. We returned individuals to their social group within 1–2 h.

Morphological Data

We measured canines and body size for 13 adult males (Table I). We measured body mass with a 5-kg spring balance to the nearest 50 g. We measured body and limb length with a tape measure to the nearest 0.5 cm. We measured the length and width of canines and testes with 150-mm digital calipers (SPI-DigiMax) to 0.1 mm. We photographed male chests with a 35-mm camera positioned 1 m from the sedated male. We subsequently used the photographs to verify categorization of the males as clean, stained, and intermediate. We calculated body mass index (BMI) as body mass (kg)/m2 tail-crown length (cf. Setchell and Dixson, 2001a). We converted testes measurements to mass via the conversion of Harcourt et al. (1995): mass (g) = 2 *volume (cm3) *1.1, wherein volume = 4/3 *π*l/2 *b1/2 * b2/2 (and l is length and b1, 2 are widths).

Behavioral Data

Lewis and 2 assistants collected behavioral data in 30-min focal individual samples (Altmann, 1974) on each adult of 6 focal groups from November 2000 to March 2002, for a total of 2875 focal h of observation. We recorded observations of social behavior—including aggression, submission, and scent-marking rates—continuously, based on a previously published ethogram for the species (Brockman, 1994). We conducted instantaneous sampling with 3-min intervals to record distances of all group members from the focal individual. Because scent marking is a rare behavior, we recorded in addition all occurrences of scent marks and scent-mark style—gouge, throat, anogenital—for 7 mo (September 2001–March 2002), during which we observed all groups for similar amounts of time. We easily recorded all scent marking by focal individuals. We did not consider urination and defecation scent marks unless one of the scent glands was rubbed against a substrate during the process. The term scent mark refers to any type of scent mark unless otherwise stated (Lewis, 2005, 2006). Behavioral data were primarily limited to males from groups A, B, E, H, and T owing to the limited data for group J.

Analysis

Because not all information was available on all males, the number of males included in an analysis varied depending on the available data. Unless otherwise stated, the sample contained 4 clean males and 7 stained males. We used scent-mark frequencies to evaluate differences in the occurrence of scent marks and throat marks because clean males scent marked so infrequently that proportions were not very informative (Lewis, 2005).

We used 1-tailed Mann-Whitney U tests to compare the 2 categories of males when stained males are expected to be larger. We used 2-tailed tests when we were testing no a priori hypothesis. Statistical analyses utilized the SPSS 11.0 package, except for the χ2 test, which we calculated by hand per Siegel and Castellan (1995). Because sample sizes are small, we calculated Cohen’s d to determine the effect of size (Cohen, 1988) with pooled standard deviation (Ronow and Rosenthal, 1996) when results were not significant. We defined a large effect size as Cohen’s d-value of d=0.8 (cf. Cohen, 1988). Whiskers on graphs represent 5th/95th percentiles; boxes give the median and 25th/75th percentiles.

RESULTS

Male Categories and Behavior

Of the 13 adult males captured (Table I), some had chests that were stained brown and greasy (Fig. 1a), whereas others had chests that were as clean and white as a female’s chest (Fig. 1b). We easily divided males into 2 distinct categories: stained (n=7) and clean (n=4). Two males (TT and TK) exhibited intermediate stages, in which a male’s chest was stained but no longer greasy, or greasy but not yet stained. TT was a solitary male that we rarely observed. TK is an interesting and informative case. We excluded both from the analyses.
Fig. 1.

(a) A sedated adult male with a stained chest. (b) A sedated adult male with a clean chest.

Fig. 2.

Scent-mark rates based upon focal data, nclean=4, nstained=6, p=0.01.

Based on ad libitum data of all scent marks, clean-chested males scent marked significantly less frequently than stained males did (Mann Whitney: Z =−2.558, n1=6, n2=4, p=.001, medianstained=187, range=72; medianclean=29, range=33). We examined the subset of the total scent marks that included ≥1 throat mark to determine the role of throat-marking in chest-staining. Males with clean chests also scent-marked with their throat glands significantly less frequently (Mann Whitney: Z =−2.745, nstained=6, nclean=4, p=.004, medianstained=79, range=56; medianclean=14, range=14). Continuous focal individual data also revealed lower scent-mark rates in clean-chested males than in stained males (Fig. 2: Mann Whitney: Z =−2.558, nstained=6, nclean=4, p=.01).

Chest Status and Group Membership

During the January 2002 census—supplemented with observations from intergroup encounters—we recorded the chest status of the males in 14 groups (Table II). We observed 14 stained males, 6 clean-chested males, and 2 males with intermediate chests. Thus the probability of a male being a resident adult clean male in the population is ca. 27%. We observed 2 solitary adult males in January: 1 (KB) had a stained chest and the other (TT) had a chest that was intermediate. With inclusion of the 2 males, the probability of being an adult clean male in the population was 25%. The distribution of male classes within social groups with respect to dominance is significantly different from random (Table III: χ2=25.28, df=3, p<.005). The number of adult males in a social group was not correlated with the number of adult females in a group (Spearman: n=17, rs=.300, p=.241). The distribution of males in a group in relation to the number of females in the group for the 17 groups we censused in January 2002 is in Fig. 3.
Table II.

Distribution of stained chests across social groups in January 2002

  

Chest status

Group

Male

Stained

Intermediate

Clean

A

LZ

x

  
 

BG

  

x

B

BN

x

  

C

TS

x

  
 

BL

  

x

E

DW

x

  
 

DP

  

x

 

AN

  

x

F

JU

x

  

G

JP

x

  
 

VN

  

x

H

SP

x

  

J

PY

x

  
 

TK

 

x

 
 

JZ

  

x

L

BB

x

  

M

BY

x

  
 

LL

 

x

 

P

SM

x

  

Q

OE

x

  

S

IZ

x

  

T

BO

x

  

Solitary

TT

 

x

 
 

KB

x

  

Note. Males in italics are the dominant males in the group based on unidirectional submissive signals or scent-marking behavior or both.

Table III.

Chest status in relation to social status

 

Clean

Stained

α-Male or only male in group

0

14

Subordinate male

8

0

All groups had only 1 stained male, while additional males had either clean or (occasionally) intermediate chests. For 4 mo, however, 2 adult males in group J had stained chests. CH was the stained male in group J when he and 2 other males (TP and JZ) became members of the group in a takeover in January 2001. On May 11 2001, TP also had a stained chest. TP remained in J with a stained chest until he disappeared in September 2001.

Males with chests of intermediate coloration sometimes resided in the same group as a stained-chested male. Group M contained 2 fully adult males of ≥5 yr (BY, LL) and 2–3 males that were potentially sexually active aged 3–4 yr (NJ, NG, BW). At the beginning of the study, LL had a stained chest and BY was clean chested. Toward the end of the study, however, their chest status was reversed.
Table IV.

Aggression matrices for the multimale groups

 

Receiver

 

LZ

BG

DW

DP

AN

Group A

     

LZ

27

   

 BG

8

   

Group E

     

DW

  

2

6

 DP

  

5

6

 AN

  

0

3

Note. Italics indicate a stained male.

Table V.

Submissive chatter matrices for the multimale groups

 

Receiver

Initiator

LZ

BG

DW

DP

AN

Group A

     

LZ

3

   

 BG

60

   

Group E

     

DW

  

0

0

 DP

  

164

1

 AN

  

119

50

Note. Italics indicate a stained male.

Chest Status and Social Behavior

Chest status in males residing in multimale groups related to dominance status. Matrices of aggressive behaviors (including supplants) and submissive chatters are in Tables IV and V. Male-male aggression rates were low (averaging 1.2/h when ≤1 m) and not unidirectional. Aggression by subordinate males toward the dominant males often occurred simultaneously with submissive signals. The submissive chatter (cf. Brockman et al., 2001) was more frequent than aggression, often spontaneously emitted, and unidirectional. The chatter data clearly indicate that the stained male in multimale groups was the dominant male.
Fig. 3.

Distribution of (a) adult males in relation to adult females and (b) adult and subadult males in relation to adult and subadult females in a group for the 17 groups censused in January 2002. Numbers and circle sizes refer to the number of groups with this distribution, rs=.300, p=.241.

Using focal individual time point samples, we examined the percentage of time that a male-female dyad was ≤10 m of each other (Fig. 4). We excluded group J males because of the limited number of focal individual samples. Thus, we analyzed 5 stained males and 3 clean males in 12 possible dyads with females. Stained-chested males spent more time ≤10 m of a female than clean-chested males did (Mann Whitney: Z =−1.868, nstained=7, nclean=5, p=0.037). Because some individuals contributed to >1 dyad, the data are not statistically independent. Pooling data for each individual male showed the same trend (Mann Whitney: Z =−1.743, nstained=5, nclean=2, p=.081, Cohen’s d=2.4). The effect size is very large, suggesting that with a larger sample size, the result would probably be statistically significant.
Fig. 4.

Relationship between chest status and time males are in association with females, nclean=3, nstained=5, ndyad=12, p<.036.

Chest Status and Morphology

Chest status was not related to body mass (1-tailed Mann Whitney: Z =−0.287, p=.394; medianstained=3.1 kg, range=0.95; medianclean=3.15 kg, range=200, Cohen’s d=.0). Contrary to expectations based on dominance status (cf. Lewis 2002), males with stained chests differed from males with clean chests in having lower body mass indices (1-tailed Mann Whitney: Z =−1.89, p=0.037; medianstained=19.22, range=4.16; medianclean=21.09, range=3.03) even though we found no difference in tail-crown length (2-tailed Mann-Whitney: Z =−1.614, p=.109; medianstained=41.0 cm, range=4; medianclean=39.3 cm, range=3, Cohen’s d=1.1).

Stained males did not exhibit significantly larger mean maxillary canine lengths (2-tailed Mann-Whitney: Z =−1.136, p=.315, Cohen’s d=.9). The mean length for the 11 males sampled is 8.7 mm. Values range from 7.4 to 10.7 mm in average upper canine length and statistical power is low (17%). Individual upper canines range from 7.3 to 11.5 mm.

The estimated absolute testes mass of the 2 classes of males does not differ significantly, though we found a trend for larger values for stained males (1-tailed Mann Whitney: Z =−1.512, p=.082; medianstained=2.20 g, range=1.34; medianclean=1.68, range=0.97, Cohen’s d=0.9). However, relative testes mass (testes mass/body mass) was not larger for stained males (1-tailed Mann Whitney: Z =−1.323, p=.115; medianstained=.072 g, range=.058; medianclean=.055 g, range=.035, Cohen’s d=.9).

Chest Status Reversibility

Male chest staining is reversible. TK was initially stained-chested while living in group J, where he was never observed to direct submissive chatters at the other adult male, ON, but received them from him. Hence, TK was the dominant male of group J at the time. However, TK and ON were overthrown by a trio of group Q males (CH, TP, JZ) in January 2001, after which TK varied from being a floater male (male without a social group) to resident male of group J to floater male. He was captured 5 times. At the time of the capture for this study, he was a peripheral resident male. His chest was not entirely clean and white, nor was it as stained and brown as other males categorized as stained; hence, the classification as intermediate.

A similar change in chest status coincident with a change in dominance status probably occurred with males LL and BY in group M. M was not a focal group but frequently interacted with focal groups J and T. Though we have no data on agonistic interactions among the 2 males, they changed roles in an important behavior that is often associated with the dominant position in the group (Lewis, 2005): frequent and immediate overmarking of female scent marks. At the beginning of the study, LL was the stained male and frequently overmarked the female’s scent marks, whereas BY was clean chested and did not overmark female scent marks. Toward the end of the study, however, BY rather than LL exhibited the behavior and had a stained chest, whereas LL had become clean chested.

Chest Status and Mating Behavior

While mating behavior occurred during both the 2001 and 2002 mating seasons, we observed behavioral estrus for only 4 females. Mating behavior varied and thus we could not pool quantitative data. The observations are described in detail in Lewis (2004), are in concordance with previous studies of Verreaux’s sifaka (Brockman, 1999; Richard, 1974, 1992), and qualitatively summarized here.

Males exhibited mate guarding (cf. Brockman, 1999) on the days when females mated. Stained males generally attempted to keep both resident and extragroup males away from the estrous female. One clean male mate guarded when the stained male disappeared from the group for a short period (a rare occurrence). No other clean male mate guarded. Guarding behavior was both aggressive and nonaggressive (cf. Brockman, 1999; Richard, 1992). Clean males either attempted to mate when they were out of sight of the stained male or simply openly persisted throughout the day forcing the stained male to make repeated choices between feeding and guarding. Influxes of males sometimes appeared when a female was in behavioral estrus. The extragroup males positioned themselves at the periphery of the group and emitted hoo vocalizations, presumably to alert the female to their presence. Resident males sometimes formed coalitions to attempt to keep them and neighboring group males from mating with the female. No intergroup matings were observed (cf. Lawler, 2003). Aggressive defense of females was not a common a strategy used against resident males. Instead of chasing resident males from estrous females, 1 strategy used by a guarding male was simply to approach the pair, place his hand on the back of the male attempting to copulate with the female, or attempt to squeeze his body between the male and female. Because the guarding male was often the stained male, the subordinate clean male would respond to his actions with a submissive signal and retreat. Copulations with a clean male were often, but not always, interrupted by another individual—often the stained male.

Males did not attack estrous females. Instead, females were sometimes aggressive toward the males. Occasionally a stained male used aggression toward a resident clean male that was attempting to copulate with the female. She responded with aggression toward the stained male. Females facilitated copulation with both stained and clean males.

DISCUSSION

Two Categories of Males

This study demonstrates that male Verreaux's sifaka in the Kirindy population can clearly be divided into two categories: clean-chested and stained-chested, with rare, intermediate forms that turned out to be in transition. Staining of the chest is a good predictor of male behavior. Stained males were dominant and exhibited intense mate guarding. The evidence for transitions of chest status in both directions indicates that the bimorphism of male Verreaux’s sifaka is more like the reversible bimorphism in mandrills than the irreversible bimorphism in orangutans. Consistent with the conclusion is the lack of striking morphological differences: If clean chest were a maturational state preceding stained chest, some morphological differences are to be expected. Long-term studies are needed to determine whether every male achieves a stained chest at least once in his lifetime or whether only a portion of the population achieves the stained status and whether a male can achieve the stained status more than once. Our preliminary results suggest that chest status in Verreaux’s sifaka represents alternative mating tactics and that a stained chest is a badge of status.

Scent Marking

Clean males scent marked significantly less frequently than stained males did. Staining on the chest probably results from the fact that stained males scent marked with their throat glands at an 8-fold higher frequency. The greasiness of the chest derives from the activity of the sternal gland. The stained aspect of the signal comes from the actual scent-marking behavior, where the throat and chest are rubbed up against a substrate (often multiple times within a single scent mark). Approximately half the scent marks by stained males are overmarks, in which a scent mark is placed on or near another scent mark (Lewis, 2005), and thus the staining is probably a combination of a male’s own glandular secretions, female anogenital glandular secretions, female urine, and dirt. Though males cannot easily autogroom the upper portion of their chests and throat, the staining usually occurs low enough on the chest that males could feasibly clean their matted and dirty hair. Moreover, individuals rarely groomed the chest during allogrooming (Lewis et al., 2003).

Distribution of Males, Rank, and Competition

The number of males in sifaka groups did not relate to the number of females in the group, consistent with the finding of Richard et al. (1993) of no systematic relationship between a group’s size and its sex ratio in the Beza Mahafaly population. Some groups had >1 adult resident male and ≤4 resident males when we included subadults. However, we found a single stained male in each social group, and when we observed additional males in groups, they were all clean chested or in the process of becoming clean chested. Exceptions appear to be the result of instability due to dominance change or perhaps migration. However, long-term data are needed to clarify the role of group instability in male chest status.

In multimale Verreaux’s sifaka groups, the activity of the sternal gland and scent marking that results in the staining of the chest appears to be related to dominance rank (cf. Kraus et al., 1999, but note that they included subadult males in their analyses). Accordingly, clean-chested males were always subordinate to the stained males. The observation that TK lost the staining on his chest after he was no longer the dominant male of a group suggests that that when a male loses α status, the staining on its chest decreases. Moreover, the observation of BY and LL switching both chest status and their roles in overmarking behavior reinforces the suggestion that a change in chest staining co-occurs with a change in rank.

Clean males spent less time in association with females than stained males did. When females were in behavioral estrus, stained males attempted to mate guard. Nevertheless, both clean males and stained males mated with females. Sifaka males cannot control mating (cf. Richard, 1992), and females exhibited aggression toward males that attempted to limit their mate choice (Lewis, 2004). The behavior contrasts with the situation in haplorrhines, in which sexual dimorphism enables males to exert greater control over female mating (van Schaik et al., 2004).

Body Size

Because chest status is reversible, more permanent morphological features, e.g., canine size and body length, are not expected to vary by chest status. Our results do not reveal a clear picture of the relationship between variable body size measurements and chest status, perhaps owing to the small sample size. However, other bimorphic species also exhibit little to no body size differences between male morphs, e.g., mandrills (Setchell, 2003). Moreover, recent findings that sifaka experience greater directional selection on limb shape than body mass (Lawler et al., 2005) suggests that the measures may not be the best indicators of male competitive ability and may help to explain our finding that clean-chested males had lower body mass indices than those of stained males. Other researchers have found rank-related differences in testes mass in Verreaux’s sifaka (Brockman et al., unpublished data). Our finding of only a trend for testes mass to be larger in stained males may be due to the small sample size.

Possible Interpretations

Dominant male sifaka have higher testosterone levels than subordinate males (Brockman et al., 1998, 2001; Kraus et al., 1999). In fact, researchers have suggested that dominant male sifaka induce the inhibition of testicular endocrine function in subordinate males via chronic stress or pheromones emitted during scent marking (Kraus et al., 1999). While we did not measure testosterone levels, we noted that stained males are dominant. It therefore follows that stained males probably also have significantly higher testosterone levels. Chest staining may thus be a signal of high testosterone to other animals, especially if sternal gland function is stimulated by testosterone (cf. Dixson, 1976). The interpretation is in accord with the finding of Kraus et al. (1999) that dominant males scent mark much more than subordinates do.

Visual Signal as a By-Product or Badge?

One explanation for the presence of chest staining as a visual signal is that the staining is a by-product of throat marking and not actually a signal at all. In fact, some male Propithecus with dark chests, such as Milne-Edward’s sifaka (P. diadema edwardsi: Summer J. Arrigo-Nelson, pers. com.) and Coquerel’s sifaka (P. verreauxi coquereli: pers. obs.), also have a stickiness of the chest hair that is produced by secretions of the sternal gland similar to that seen in stained males of Propithecus subspecies with light-colored chests (e.g., P. v. verreauxi and P. diadema candidus: Erik Patel, pers. com.). Because of pigmentation of the chest hair, the secretions produce a less conspicuous visual signal in the subspecies with dark-pigmented chests. Though it is not currently known whether individuals use chest staining as a visual means to evaluate male chest status, it seems unlikely that individuals would overlook the valuable information provided by the staining. Thus, staining of the chest may have started as a by-product of olfactory communication (scent marking with the sternal gland), but the resultant visual signal is unlikely to be ignored.

An alternative, and more likely, explanation is that Verreaux’s sifaka have evolved a visual signal of dominance status (i.e., fighting ability, cf. Lewis, 2002) that probably reflects testosterone levels. Male mandrills exhibit variation in male sternal gland activity that is rank dependent (Setchell and Dixson, 2001a,b; 2002). Changes in dominance status in adult mandrill males results in only some changes in morphological features. Male mandrills that gain α status develop greater testicular volume, testosterone, reddening of the sexual skin, and fatted rumps, but not an increase in body mass or blue coloration (Setchell and Dixson, 2001a). Similarly, males losing α status experience a reduction of most of the features, suggesting that the crucial variable in determining morphological variance in male mandrills is testicular function (Setchell and Dixson, 2001b).

Mandrill social groups are generally enormous (≥600 individuals) and have highly skewed sex ratios (Abernathy et al., 2002). Because males and females are unlikely to have good previous knowledge of other individuals, signals are important for communication (Setchell and Wickings, 2005), like badges in birds (Preuschoft and van Schaik, 2000). Male and female Verreaux’s sifaka associate in small, permanent social groups. However, group composition is not always stable (Brockman et al., 2001; Lewis et al., 2003), and “[s]patial boundaries of groups do not coincide with social or reproductive boundaries…” (Richard, 1985, p. 553). Home ranges of social groups at Kirindy overlap an average of 77% (Lewis, 2004), and a large number of Verreaux’s sifaka males reproduce with females outside of their social group (Lawler, 2003). These characteristics led Jolly (1966) and Richard (1978) to suggest that neighborhoods are important for understanding Verreaux’s sifaka social behavior. Thus, visual signals may be also useful for intergroup, intraneighborhood communication. A visual signal would allow neighboring group members to identify males of different dominance status (and presumably testosterone levels) quickly. Experiments have shown that scent marks operate only over short distances in thick-tailed galagos (Otolemur crassicaudatus: Katsir and Crewe, 1980); therefore, visual signals may be more important than previously thought for strepsirrhines, especially during face-to-face encounters (Mertl, 1976).

Male sifaka are the dispersing sex, though females also sometimes disperse (Kubzdela, 1997; Richard et al., 1993). Most males transfer to neighboring groups (Richard et al., 1993; Lewis et al., 2003), but solitary males, and occasionally male pairs, pass through the home ranges of other groups. Long-distance dispersal events also occur (Richard et al., 1993; Lewis et al., 2003), and visual signals may be useful for these cases. Migrating males can be clean, stained, or intermediate. Testosterone levels generally increase when a male transfers groups aggressively (Brockman et al., 2001). Thus, chest status may signal to the resident male(s) the strategy of a migrating male if he tries to join their group: takeover or join as a subordinate. In fact, Verreaux’s sifaka males react differently depending on the chest status of another male during encounters with extragroup males (R. J. Lewis, pers. obs.).

Male vervets (Cercopithecus aethiops) also exhibit a reversible signal of status despite living in mixed-sex social groups. In an elegant experimental study, Gerald (2001) demonstrated the influence of scrotal coloration in signaling dominance status. She posited that transferring males may utilize the visual signal to gather information about the potential for aggression from resident males. Scrotal coloration in vervets correlates more with serotonin than with testicular function (Gerald and Raleigh, 1997); nevertheless, scrotal coloration in vervets and chest status in sifaka may both be examples of badges in primates (cf. Gerald, 2001, 2003).

Alternative Mating Tactics

The fact that every group contains 1 stained male begs the question of why some groups have ≥1 clean males in addition to the stained male. Clean-chested males made up only a quarter of the population in our study. One possibility is that stained males may simply not tolerate other stained males. By maintaining low testosterone levels, or reducing testosterone levels, clean males may make themselves less threatening to the stained males. Subordinate male birds sometimes mimic female plumage or reduce badge size to be less threatening to territory-holding males (Rohwer et al., 1980; Slagsvold and Sætre, 1991). Similarly, unflanged male orangutans are tolerated by the flanged males (Utami Atmoko and van Hooff, 2004). With a clean chest, less active sternal gland, and smaller testes, the clean-chested male sifaka may be employing a similar strategy. Long-term data are needed to determine whether the proportion of clean-chested males in the population is density dependent, like unflanged males in Sumatran orangutans (Utami Atmoko and van Hooff, 2004).

Male group tenure in Verreaux’s sifaka lasts an average of 5 yr (Richard et al., 1993). Thus, most males will find themselves as adults trying to enter a new social group that already has a stained male. One strategy that males can use is to fight their way into the group and enter at the top of the male hierarchy as a stained male, but alternatively males may reduce their testosterone and enter the group as subordinate clean males (cf. van Noordwijk and van Schaik, 1985). Because clean males were able to copulate with females in their group, having a clean chest may represent an alternative mating tactic, in which males self-inhibit testosterone levels. Indeed, male polymorphisms are expected in species that exhibit high variance in male reproductive success (Shuster and Wade, 2003), like Verreaux’s sifaka (Lawler, 2003). Setchell and Dixson (2001a) also suggest that the different male morphs may represent alternative mating tactics in mandrills. Nonfatted males may reduce their investment in costly secondary sexual adornments, avoid competition within groups, and sneak matings opportunistically, perhaps ultimately resulting in similar reproductive success (Setchell, 2003; Setchell and Dixson, 2001a).

Being a clean-chested male seems to be a viable strategy for males. Clean males reside in permanent social groups, and thus benefit from added protection from predation. Resident females that are in estrus also seek out copulations with resident clean males. Moreover, the costs of residing in social groups with a stained male are presumably not high because intragroup intrasexual aggression is rare (Table IV). Therefore, an adult male that is not a stained male of a social group may be better off to use the clean chest strategy than to be a solitary male.

Conclusion

Like the bimorphism in male mandrills (Setchell and Dixson, 2001a,b), chest status in Verreaux’s sifaka at Kirindy may signal alternative tactics males use to maximize reproductive success based on their social environment. If a male has no opportunity to attempt to guard a female and reproduce as a stained male, then he can use the tactic of being unobtrusive, lowering testosterone levels, and sneaking matings when possible. While some researchers have suggested subordinate males are forced into the strategy by the stained male suppressing their testosterone levels (Kraus et al., 1999), the observed differences in males may actually be alternative tactics. Maintaining clean-chest status may be a way for some males to increase their reproductive success by being tolerated in social groups while they bide their time until they too can become a stained male. The clean-chest strategy may also facilitate multimale, multifemale social groups by increasing social stability with a visual badge of status. Experimental tests and paternity data (currently under investigation) are needed to understand fully the phenomenon of bimorphism in Verreaux’s sifaka.

Notes

Acknowledgments

We thank D. Brockman, M. Gerald, K. Glander, P. Kappeler, C. Kirk, K. Smith, M. Munger, D. Overdorff, and J. Mitani for valuable discussion and comments on the manuscript as well as 4 anonymous reviewers. Discussion in a departmental seminar at Duke University greatly improved the research. We thanks P. Kappeler, DPZ, the Malagasy Government, Eaux et Fôrets, CFPF, MICET, D. Rakotondavony, Enafa, J. Ratsirarson, and the Kirindy sifaka team for assistance in captures. Research methods were in compliance with the requirements of Madagascar. The Wenner-Gren and Leakey Foundations funded the research. The material is based on work supported by the National Science Foundation Dissertation Improvement Grant to RJL under grant no. 0002570.

REFERENCES

  1. Abernathy, K. A., White, L. J. T., and Wickings, E. J. (2002). Hordes of mandrills (Mandrillus sphinx): Extreme group size and seasonal male presence. J. Zool. Lond. 258: 131–137.Google Scholar
  2. Altmann, J. (1974). Observational study of behavior: Sampling methods. Behaviour 48: 227–265.Google Scholar
  3. Brockman, D. K. (1994). Reproduction and Mating System of Verreaux’s Sifaka, Propithecus verreauxi, at Beza Mahafaly, Madagascar. Doctoral thesis. Yale University, New Haven, CT.Google Scholar
  4. Brockman, D. K. (1999). Reproductive behavior of female Propithecus verreauxi at Beza-Mahafaly, Madagascar. Int. J. Primatol. 20: 375–398.CrossRefGoogle Scholar
  5. Brockman, D. K., Whitten, P. L., Richard, A. F., and Schneider, A. (1998). Reproduction in free-ranging male Propithecus verreauxi: The hormonal correlates of mating and aggression. Am. J. Phys. Anthropol. 105: 137–151.PubMedCrossRefGoogle Scholar
  6. Brockman, D. K., Whitten, P. L., Richard, A. F., and Benander, B. (2001). Birth season testosterone levels in male Verreaux’s sifaka, Propithecus verreauxi: Insights into socio-demographic factors mediating seasonal testicular function. Behav. Ecol. Sociobiol. 49: 117–127.CrossRefGoogle Scholar
  7. Cohen J. (1988). Statistical Power Analysis for the Behavioral Sciences, 2nd ed. Lawrence Earlbaum, Hillsdale, NJ.Google Scholar
  8. Dawkins, R. (1980). Good strategy or evolutionary stable strategy? In Barlow, G. W., and Silverberg, J. (eds.), Sociobiology: Beyond Nature/Nurture? Westview, Boulder, CO, pp. 331–367.Google Scholar
  9. Dill, A. (2000). Aspekte der Jungenaufzucht bei Larvensifakas (Propithecus verreauxi verreauxi, Grandidier 1867, Primates: Indriidae). Msc-thesis. University of Würzburg, Würzburg, Germany.Google Scholar
  10. Dixson, A. F. (1976). Effects of testosterone on the sternal cutaneous glands and genitalia of the male greater galago (Galago crassicaudatus crassicaudatus). Folia Primatol. 26: 207–213.PubMedCrossRefGoogle Scholar
  11. Dixson, A. F. (1998). Primate Sexuality: Comparative Studies of the Prosimians, Monkeys, Apes, and Human Beings. Oxford University Press, Oxford.Google Scholar
  12. Ganzhorn, J. U. (2002). Distribution of a folivorous lemur in relation to seasonally varying food resources: Integrating quantitative and qualitative aspects of food characteristics. Oecologia 131: 427–435.CrossRefGoogle Scholar
  13. Ganzhorn, J. U., and Sorg, J.-P., eds. (1996). Ecology and economy of a tropical dry forest in Madagascar. Primate Report 46–1.Google Scholar
  14. Gerald, M. S. (2001). Primate colour predicts social status and aggressive outcome. Anim. Behav. 61: 559–566.CrossRefGoogle Scholar
  15. Gerald, M. S. (2003). How color may guide the primate world: Possible relationships between sexual selection and sexual dichromatism. In Jones, C. B. (ed), Sexual Selection and Reproductive Competition in Primates: New Perspectives and Directions. American Society of Primatologists, Norman, OK, pp. 141–171.Google Scholar
  16. Gerald, M. S., and Raleigh, M. J. (1997). A potential proximate mechanism for scrotal colour variation in captive vervet monkeys (Cercopithecus aethiops sabaeus). Am. J. Primatol. 42: 111–112.Google Scholar
  17. Glander, K. E., Wright, P. C., Daniels, P. S., and Merenlender, A. M. (1992). Mophometrics and testicle size of rain forest lemur species from southeastern Madagascar. J. Hum. Evol. 22: 1–17.CrossRefGoogle Scholar
  18. Grieser, B. (1992). Infant development ad parental care in two species of sifakas. Primates 33: 305–314.CrossRefGoogle Scholar
  19. Jolly, A. (1966). Lemur Behavior: A Madagascar Field Study. University of Chicago Press: Chicago, IL.Google Scholar
  20. Katsir, Z., and Crewe, R. M. (1980). Chemical communication in Galago crassicaudatus: investigation of the chest gland secretion. S. Afr. J. Zool. 15: 249–254.Google Scholar
  21. Kraus, C., Heistermann, M., and Kappeler, P. M. (1999). Physiological suppression of sexual function of subordinate males: A subtle form of intrasexual competition among male sifakas (Propithecus verreauxi)? Physiol. Behav. 66: 855–861.PubMedCrossRefGoogle Scholar
  22. Kubzdela, K. S. (1997). Sociodemography in Diurnal Primates: The Effects of Group Size and Female Dominance Rank on Intra-group Spatial Distribution, Feeding Competition, Female Reproductive Success, and Female Dispersal Patterns in White Sifaka, Propithecus verreauxi verreauxi. Doctoral thesis. University of Chicago, Chicago.Google Scholar
  23. Lawler, R. R. (2003). Causes and Consequences of Differential Reproductive Success in Male White Sifaka (Propithecus verreauxi verreauxi). Doctoral thesis. Yale University, New Haven, CT.Google Scholar
  24. Lawler, R. R., Richard, A. F., and Riley, M. A. (2005). Intrasexual selection in Verreaux’s sifaka (Propithecus verreauxi verreauxi). J. Hum. Evol. 48: 259–277.PubMedCrossRefGoogle Scholar
  25. Lewis, R. J. (2004). Male-Female Relationships in Sifaka (Propithecus verreauxi verreauxi): Power, Conflict, and Cooperation. Doctoral thesis. Duke University, Durham, NC.Google Scholar
  26. Lewis, R. J. (2005). Sex differences in scent-marking in sifaka: Mating conflict or male services? Am. J. Phys. Anthropol. 128: 389–398.PubMedCrossRefGoogle Scholar
  27. Lewis, R. J. (2006). Scent-marking in sifaka: No one function explains it all. Am. J. Primatol. 68(6): 622–636.Google Scholar
  28. Lewis, R. J., Razafindrasamba, S. M., and Tolojanahary, J. P. (2003). Observed infanticide in a seasonal breeding prosimian (Propithecus verreauxi verreauxi) in Kirindy Forest, Madagascar. Folia Primatol. 74(2): 101–103.Google Scholar
  29. Lewis, R. J., and Kappeler, P. M. (2005). Seasonality, Body Condition, and the Timing of Reproduction in Propithecus verreauxi verreauxi in the Kirindy Forest. Am. J. Primatol. 67(3): 347–364.Google Scholar
  30. Maggioncalda, A. N., Sapolsky, R. M., and Czekala, N. M. (1999). Reproductive hormone profiles in captive male orangutans: Implications for understanding developmental arrest. Am. J. Phys. Anthropol. 109: 19–32.PubMedCrossRefGoogle Scholar
  31. Maggioncalda, A. N., Czekala, N. M., and Sapolsky, R. M. (2000). Growth hormone and thyroid stimulating hormone concentrations in captive male orangutans: Implications for understanding developmental arrest. Am. J. Primatol. 50: 67–76.PubMedCrossRefGoogle Scholar
  32. McGraw, K. J., Dale, J., and Mackillop, E. A. (2003). Social environment during molt and the expression of melanin-based plumage pigmentation in male house sparrows (Passer domesticus). Behav. Ecol. Sociobiol. 53: 116–122.Google Scholar
  33. Mertl, A. S. (1976). Olfactory and visual cues in social interactions of Lemur catta. Folia Primatol. 26: 151–161.Google Scholar
  34. Petter, J.-J. (1962). Recherches sur l’ecologie et l’ethologie des lemuriens Malagaches. Mém. Mus. Nat. Hist. Nat. 27: 1–146.Google Scholar
  35. Preuschoft, S., and van Schaik, C. P. (2000). Dominance and communication: Conflict management in various social settings. In Aureli, F., and de Waal, F. B. M. (eds), Natural Conflict Resolution. University of California Press, Berkeley, pp. 77–105.Google Scholar
  36. Rensing, S. (1999). Immobilization and anesthesia of nonhuman primates. Primate Report 55: 33–38.Google Scholar
  37. Richard, A. F. (1978). Behavioral Variation: Case Study of a Malagasy Lemur. Bucknell University: Lewisburg, PA.Google Scholar
  38. Richard, A. F. (1985). Social boundaries in a Malagasy prosimian, the sifaka (Propithecus verreauxi). Int. J. Primatol. 6: 553–568.Google Scholar
  39. Richard, A. F. (1992). Aggressive competition between males, female-controlled polygyny and sexual monomorphism in a Malagasy primate, Propithecus verreauxi. J. Hum. Evol. 22: 395–406.CrossRefGoogle Scholar
  40. Richard, A. F., Rakotomanga, P., and Schwartz, M. (1993). Dispersal by Propithecus verreauxi at Beza Mahafaly, Madagascar: 1984–1991. Am. J. Primatol. 30: 1–20.CrossRefGoogle Scholar
  41. Richard, A. F., Dewar, R. E., Schwartz, M., and Ratsirarson, J. (2000). Mass change, environmental variability and female fertility in wild Propithecus verreauxi. J. Hum. Evol. 39: 381–391.PubMedCrossRefGoogle Scholar
  42. Richard, A. F., Dewar, R. E., Schwartz, M., and Ratsirarson, J. (2002). Life in the slow lane? Demography and life histories of male and female sifaka (Propithecus verreauxi verreauxi). J. Zool. Lond. 256: 421–436.Google Scholar
  43. Rohwer, S., and Ewald, P. W. (1981). The cost of dominance and advantage of subordination in a badge signaling system. Evolution 35: 441–454.CrossRefGoogle Scholar
  44. Rohwer, S., Fretwell, S. D., and Niles, D. M. (1980). Delayed maturation in passirine plumages and the deceptive acquisition of resources. Am. Nat. 115: 400–437.CrossRefGoogle Scholar
  45. Ronow R. L., and Rosenthal R. (1996). Computing contrasts, effect sizes, and countermills on other people’s published data: General procedures for research consumers. Psychol. Methods 1: 331–340.CrossRefGoogle Scholar
  46. Schilling, A. (1979). Olfactory communication in prosimians. In Doyle, G. A., and Martin, R. D. (eds), The Study of Prosimian Behavior. Academic Press, New York, pp. 461–542.Google Scholar
  47. Setchell, J. M. (2003). The evolution of alternative reproductive morphs in male primates. In Jones, C. B. (ed), Sexual Selection and Reproductive Competition in Primates: New Perspectives and Directions. American Society of Primatologists: Norman, OK, pp. 271–281.Google Scholar
  48. Setchell, J. M., and Dixson, A. F. (2001a). Arrested development of secondary sexual adornments in subordinate adult male mandrills (Mandrillus sphynx). Am. J. Phys. Anthropol. 115: 245–252.PubMedCrossRefGoogle Scholar
  49. Setchell, J. M., and Dixson, A. F. (2001b). Changes in secondary sexual adornments of male mandrills (Mandrillus sphynx) are associated with gain and loss of alpha status. Horm. Behav. 39: 177–184.PubMedCrossRefGoogle Scholar
  50. Setchell, J. M., and Dixson, A. F. (2002). Developmental variables and dominance rank in adolescent male mandrills (Mandrillus sphynx). Am. J. Primatol. 56: 9–25.PubMedCrossRefGoogle Scholar
  51. Setchell, J. M., and Wickings, J. (2005). Dominance, status signals, and coloration in male mandrills (Mandrillus sphinx). Ethology 111: 25–50.CrossRefGoogle Scholar
  52. Siegel, S., and Castellan, N. J. (1988). Nonparametric Statistics for the Behavioral Sciences. McGraw-Hill, New York.Google Scholar
  53. Shuster, S. M., and Wade, M. J. (2003). Mating Systems and Strategies. Princeton University Press, Princeton, NJ.Google Scholar
  54. Slagsvold, T., and Sætre, G.-P. (1991). Evolution of plumage colour in male pied flycatchers (Ficedula hypoleuca): Evidence for female mimicry. Evolution 45: 910–917.CrossRefGoogle Scholar
  55. Utami, S. S., Goossens, B., Bruford, M. W., de Ruiter, J. R., and van Hooff, J. A. R. A. M. (2002). Male bimaturism and reproductive success in Sumatran orang-utans. Behav. Ecol. 13: 643–652.CrossRefGoogle Scholar
  56. Utami Atmoko, S., and van Hooff, J. A. R. A. M. (2004). Alternative male reproductive strategies: male bimaturism in orangutans. In Kappeler, P. M., and van Schaik C. P. (eds), Sexual Selection in Primates: New and Comparative Perspectives. Cambridge University Press, Cambridge, pp. 196–207.Google Scholar
  57. van Noordwijk, M. A., and van Schaik, C. P. (1985). Male migration and rank acquisition in wild long-tailed macaques (Macaca fascicularis). Anim. Behav. 33: 849–861.CrossRefGoogle Scholar
  58. van Schaik, C. P. (2004). Among Orangutans: Red Apes and the Rise of Human Culture. Harvard University Press, Cambridge, MA.Google Scholar
  59. van Schaik, C. P., Pradhan, G. R., and van Noordwijk, M. A. (2004). Mating conflict in primates: infanticide, sexual harassment and female sexuality. In Kappeler, P. M., and van Schaik, C. P. (eds), Sexual Selection in Primates: New and Comparative Perspectives. Cambridge University: Cambridge, pp. 131–150.Google Scholar

Copyright information

© Springer Science+Business Media, LLC 2007

Authors and Affiliations

  1. 1.Department of AnthropologyUniversity of Texas at AustinAustinUSA
  2. 2.Anthropological Institute and MuseumUniversity of Zürich8057 ZürichSwitzerland

Personalised recommendations