Contexts of rubbing behavior in Alouatta guariba clamitans a scent-marking role.код для вставкиСкачать
American Journal of Primatology 70:575–583 (2008) RESEARCH ARTICLE Contexts of Rubbing Behavior in Alouatta guariba clamitans: A Scent-Marking Role? ZELINDA MARIA BRAGA HIRANO, ISABEL COELHO CORREA, AND DILMAR ALBERTO GONC- ALVES DE OLIVEIRA Centro de Pesquisas Biológicas de Indaial, Departamento de Cieˆncias Naturais, Universidade Regional de Blumenau, Blumenau, Brazil Rubbing behaviors are well known in several primate species and are usually seen as scent-marking behaviors, with several functions proposed but still widely debated. The genus Alouatta is highly sexually dimorphic and a suitable subject for the study of sexual and hierarchical divergences associated with rubbing behavior: males should mark more than females, and dominant individuals more than subordinate ones. Three wild groups of Southern brown howler monkeys, Alouatta guariba clamitans, were studied at Morro Geisler, Indaial, Brazil, from September 2004 to February 2005. One hundred and twenty-three rubbing episodes were registered; data on performers and associated contexts showed that anogenital, dorsum and hyoid regions were the most often rubbed. Adult males rubbed significantly above expected levels, whereas subordinated females and juveniles tended to rub below the expected levels. Females were the main performers of anogenital rubbing, often preceded by defecation. The predominance of rubbing in males probably serves an important function in intrasexual communication and social interactions. Intrasexual competition can also lead to a relationship between rubbing and social status in females. Hyoid and sternum rubbing by males are probably agonistic signals associated with extragroup conflict. The possible cleaning function of anogenital rubbing does not preclude a communicative function. Whether rubbing behavior in howlers is solely for the function of scent marking or can also be a visual signal (e.g. as a display or to color the substrate with pigment) c 2008 Wiley-Liss, Inc. requires further study. Am. J. Primatol. 70:575–583, 2008. Key words: Alouatta guariba; marking behavior; hierarchy; territoriality INTRODUCTION Rubbing is a well-known behavior in many mammalian species, including primates, in which a scent-marking function is frequently assumed, as body surfaces used in rubbing usually have skin glands with a putative odoriferous function [Epple, 1986]. Rubbing as a scent-marking mechanism is most studied in strepsirrhines and platyrrhines [Epple, 1986; Heymann, 2006], primate groups in which both multiple skin glands and functional accessory (vomeronasal) olfaction are found [Evans, 2006]. Chemical analyses have been carried out on primate scent marks, indicating the potential for the identification of species, sex, reproductive status and even individual signatures [Hayes et al., 2004; Smith, 2006]. Mammalian scent-marking behavior is usually related to territorial and social functions. A territorial role is assumed when marking behavior is related to intergroup encounters and home range boundaries, with marks used to signal the use of resources or area by a group [Gorman, 1990; Mertl-Millhollen, 2006]. However, when home ranges are so large as to preclude group members from the effective marking of boundaries, territorial marking can predominate r 2008 Wiley-Liss, Inc. in core areas [‘‘hinterland marking’’: Gorman, 1990; Heymann, 2000]. Social functions of skin glands are either related to hierarchic signaling, with dominant individuals using marks to reinforce their status over subordinates [Miller et al., 2003; Ralls, 1971; Walraven & van Elsacker, 1992], or sexual signaling [Di Fiore et al., 2006; Epple, 1986; Heymann, 2003a,b, 2006; Setz & Gaspar, 1997] or both functions. In Neotropical primates, rubbing behavior is well known, and a direct relationship between social rank and rubbing frequency has been found in some species [Kleiman & Mack, 1980; Miller et al., 2003; Contract grant sponsor: CNPq; Contract grant number: 305372/ 2002-5; FURB. Correspondence to: Zelinda Maria Braga Hirano, Centro de Pesquisas Biológicas de Indaial, Departamento de Ciências Naturais, Universidade Regional de Blumenau, Rua Antônio da Veiga, 140, 89102-900 Blumenau, SC, Brazil. E-mail: email@example.com Received 2 July 2007; revised 5 December 2007; revision accepted 13 January 2008 DOI 10.1002/ajp.20531 Published online 5 March 2008 in Wiley InterScience (www. interscience.wiley.com). 576 / Hirano et al. Rylands, 1990; Walraven & van Elsacker, 1992]. Sexual selection plays a role in platyrrhine scent marking, as there is a general trend for the sex having the lesser degree of parental care to mark more, which is associated with a greater level of intrasexual competition [Heymann, 2003a, 2006]. Additionally, the rubbing behavior of Neotropical primates may be related to an intersexual signaling function [Di Fiore et al., 2006; Setz & Gaspar, 1997], although a territorial function is still controversial [Gosling & Roberts, 2001; Heymann, 2000, 2006]. The described patterns of rubbing behaviors in Alouatta are assumed to serve scent-marking functions [Neville et al., 1988]. Of these behaviors, throat rubbing is usually associated with conflicts, as in intergroup encounters in Alouatta seniculus [Sekulic & Eisenberg, 1983]. Anal rubbing has been linked to social signaling in A. seniculus [Braza et al., 1981]. Neville  observed dorsal, perineal, chin, facial and neck (throat) rubbing modes in A. seniculus, but judged these behaviors to be more likely just scratching. Shoemaker  noticed the same behaviors in captive A. caraya of all ages and both sexes and considered hygienic and marking functions to be equally possible. Eisenberg , studying captive A. palliata, considered chin and chest (sternum) rubbing to be a male behavior associated with high arousal, whereas anogenital rubbing to be a female behavior without a clear context. He also described a ‘‘back-roll’’ behavior that resembles dorsal rubbing, relating it as possible artifact of captivity. Altmann  also observed genital rubbing in females of this species and suggested that it could be considered masturbation. However, neck and chest rubbing by a former solitary male A. palliata was associated with changes in his social status—access to females and group membership [Young, 1982]. Clearly, most forms of rubbing in Alouatta have not been systematically studied and are still poorly understood. Other forms of chemical communication in Alouatta involve urine washing and sniffing in A. palliata [Eisenberg, 1976; Jones, 2003; Milton, 1975] and anogenital sniffing and licking in A. palliata [Altmann, 1959; Eisenberg, 1976] and A. seniculus [Neville, 1972]. Anogenital investigation was observed to be a behavior performed by both sexes [Eisenberg, 1976] or just by males [Altmann, 1959; Neville, 1972]. Some authors also considered the acts of collective defecation in howlers as a likely scentmarking behavior, related to territoriality [Braza et al, 1981; Emmons & Feer, 1990; Shoemaker, 1979]. Recent research indicates that howler monkeys have fully functional vomeronasal sensory organs, being genetically able to perceive pheromones [Webb et al., 2004]. Even so, responses to pheromone signals are no longer considered to be exclusively dependent on vomeronasal structures, but are also Am. J. Primatol. believed to be perceived by the main olfactory epithelium, and the integration between both systems enhances the associative and learning capacities related to chemical communication in mammals [Brennan & Zufall, 2006]. Therefore, there is a wide potential for chemosignaling in Alouatta. As howlers are highly sexually dimorphic animals, we hypothesize that sexual divergence in rubbing activity can be predicted. Males should be the main performers, according to Heymann’s  schema, as they are the most competitive sex. If a hierarchical function is served by howler’s rubbing behavior, dominant individuals should mark more often than subordinate ones. We can also test hypotheses of associations between some forms of rubbing and particular contexts, based on what has been previously described in howlers. Therefore, we examine whether rubbing occurs principally in the context of intergroup conflict, particularly when the hyoid (throat) region was used, and we explore the possibility that anal rubbing is linked to defecation. To test these hypotheses, we investigated rubbing behavior in three wild groups of A. guariba clamitans. Specifically, we focused on individuals performing rubbing behavior, body parts rubbed and the contexts of rubbing behavior. METHODS Study Site This research was conducted at Morro Geisler, a forested hill located at (261540 S, 401130 W) in Indaial, Santa Catarina State, in Southern Brazil. It is a 40 ha fragment of essentially secondary growth forest that is part of the Atlantic Rainforest. Deforestation occurred at this site approximately 60 years ago, and it contains some ancient remnant trees and several clearings occupied by bamboo, grasses, lianas and chayote (Sechium edule). Height varies from 80 to 285 m above sea level. The research area is connected by forest corridors to neighboring hills still covered by similar forest as part of the hilly surroundings of the Itajai-Ac- ú River Valley. Subjects The Southern brown howler monkey, A. guariba clamitans, is the only primate species at Morro Geisler, albeit the tufted capuchin monkey, Cebus nigritus, is also a member of the native fauna of the region and can occur in neighboring areas. The local population of howlers is estimated at 30–40 individuals, with at least six groups known in the area. Three groups, named Huebes, Pedreira and Licurana, were observed from September 2004 to February 2005. The composition of the groups and the changes that occurred in these groups are given in Table I. In the two groups with two adult females, one (AF1) was probably dominant to the other (AF2) Rubbing Behavior in Howlers / 577 TABLE I. Composition of Study Groups at Morro Geisler Adult male Huebes Pedreira Licurana Population a 1 1 1 3 Adult female 1 2 2 5 Subadult male Juvenile a 1 1 2 Infant 1 2 3 1 1b 1 1–3 Total 3–4 4–5 7 14–16 Born in October 2004. Born in January 2005. b based on larger body size and the predominance of a leading position in group progression. The single female in the Huebes Group was also denominated as AF1. Each group was followed 12 hr/day for 2–4 days/ month. Thirty-two complete days of observation were obtained as follows: Huebes Group, 13 days (September 2004–January 2005); Pedreira Group, 15 days (September–December 2004, February 2005); Licurana Group, 4 days (January–February 2005). Rubbing Behavior All rubbing behavior episodes (repeated contact of any body part on a substrate) were recorded [‘‘all occurrences’’ sampling method; Altmann, 1974]. These episodes constituted single acts of rubbing by a subject. When other individuals rubbed at the same time, or the same individual rubbed another body part, it was considered as another episode. For each episode the following data were registered: hour; performer; body part rubbed (hyoid, mandible, dorsum, abdomen, sternum, anogenital region); substrate (vertical trunk, branch); context preceding the episode (intergroup conflict, defecation, play, locomotion, rain, feeding, grooming, vocalization, movement in the forest, researcher’s approach, no apparent reason); context after the episode (same as the previous ones); position of performer (sitting, lying on back, ventral contact, suspended by tail); body reactions of performer (piloerection, none); looking direction (other group members, human observers, other howler group or no target identified). To preserve mutual exclusivity of context categories, rain was considered as a context only when any other context category was not observed. Statistical Analysis The data were compiled into contingency tables and analyzed by Pearson’s w2 tests. When a significant result was found (a 5 0.05), the standardized adjusted residuals [STAR; Agresti, 2002] of each cell in a table were calculated, to show which differences were responsible for the significance. When w2 calculations involved a single column, the simpler adjusted residuals [AR; Agresti, 2002] were calculated. For an a 5 0.05, residual values higher than 1.96 indicate a frequency significantly greater than that expected in the cell; values lower than 1.96 indicate a frequency significantly below the expected one. As multiple tests are done within a single contingency table, false significant results could occur just by chance. To prevent type I errors (incorrect rejection of the null hypothesis 5 false positive), a Bonferroni-like correction was made in multiple comparisons within each contingency table. The a level is recalculated with a 5 0.05/N, where N is the number of cells in the contingency table. Therefore, a critical modular value for residuals was calculated in each w2 table, corresponding to the corrected a-value, and denominated as critical residual (CR). Residual values above positive CR or below negative CR were considered significant. We calculated the expected frequencies of rubbing episodes by each sex–age category from the number of individuals within each category in the three groups and from the time (number of days) spent with each group. Fisher exact tests were performed to show any associations between the most frequent forms of contexts (intergroup conflict and defecation) and the kind of rubbing most frequently associated with each context (respectively, hyoid rubbing and anogenital rubbing). A w2 test of adherence was used to test the hypothesis that rubbing in intergroup conflicts diverged from the expected. The time spent in intergroup conflicts by focal groups was divided by total time of observation to provide the expected proportion. Intergroup conflict duration was measured as the time from the beginning of the first loud call emission between the focal groups and other groups to the end of the last emission, in any series of successive emissions involving the same groups with less than 1 hr of interval between consecutive emissions. Other signs of intergroup conflicts, such as chasing and physical aggression, were not observed in this study. This study is in compliance with animal care regulations applicable in Brazil. All research protocols reported in this article were reviewed and approved by the Ethics Committee on Animal Experimentation of the Regional University of Blumenau at Blumenau, Brazil (Ethics Committee on Animal Experimentation protocol number 016/04). Am. J. Primatol. 578 / Hirano et al. TABLE II. Rubbing Episodes According to Body Part Rubbed in Three Groups at Morro Geisler Hyoid Sternum Anogenital Dorsum Abdomen Mandible Total 5 4 1 4 2 1 29 24 0 30 10 5 2 3 0 3 0 0 73 43 7 10 7 53 45 5 3 123 Huebes Pedreira Licurana Total TABLE III. Percentage of Rubbing Episodes (n 5 123) According to Body Part in Groups at Morro Geisler Hyoid Sternum Anogenital Dorsum Abdomen Mandible Total AM AF1 AF2 SM JU 6.5 0.0 0.0 0.0 0.8 4.9 0.8 0.0 0.0 0.0 13.0 11.4 4.9 9.8 4.1 15.4 8.9 1.6 8.9 1.6 1.6 1.6 0.0 0.0 0.8 1.6 0.8 0.0 0.8 0.0 43.1 23.6 6.5 19.5 7.3 Total 7.3 5.7 43.1 36.6 4.1 3.3 100.0 AM, adult male; AF1 and AF2, adult females; SM, subadult male; JU, juvenile. RESULTS One hundred and twenty-three episodes of rubbing behavior were recorded for the three howler groups at Morro Geisler (0.32 episode/hr). The data by group are summarized in Table II, with the Huebes Group comprising 59.5% of the rubbing events observed, followed by Pedreira Group (34.7%) and Licurana Group (5.8%). The anogenital region was the most often rubbed (43.1%), followed by the dorsum (36.6%) and hyoid (7.3%). Adult males performed rubbing behaviors more often than other group members, with 43.1% of records (Table III). The adult females were second (30.1%), followed by subadult males (19.5%) and juveniles (7.3%). The proportion that each sex–age category performed rubbing varied significantly (w2 5 24.5, df 5 3, Po0.001), with significant residuals (N 5 4, CR 5 2.50) for adult males performing above the expected (AR 5 3.59) and juveniles performing below the expected (AR 5 2.67). Dominant females performed most female rubbing episodes (AF1 5 23.6%, AF2 5 6.5%), and the separation of female data into these two categories also results in significant variation in rubbing between sex–age categories (w2 5 26.5, df 5 4, Po0.001), and adds a negative, but not significant residual to AF2 (AR 5 2.23, N 5 5, CR 5 2.58), indicating that subordinate females tend to rub below the expected. Adult males were the main performers of dorsal, hyoid and sternal rubbing, whereas adult females were the main performers of anogenital rubbing (Table III). Other forms of rubbing behaviors were infrequent, preventing a global analysis of rubbing according to body part rubbed in relation to sex–age categories. However, a partial analysis of this table, limited to the most frequent performers (adult males Am. J. Primatol. and females) and body parts rubbed (anogenital, dorsum and hyoid1sternum), showed a significant difference in the body parts most used by each sex (w2 5 10.5, df 5 2, Po0.01). The residuals (N 5 6, CR 5 2.64) indicated a significantly greater amount of hyoid and sternal rubbing by adult males (STAR 5 7.92) and a significantly greater amount of anogenital rubbing by adult females (STAR 5 3.20). Hyoid and sternal rubbing were grouped together in this analysis as they are usually performed by adult males and occur in similar contexts (see below). Substrate Rubbed Howlers only rubbed on tree branches (horizontal substrates) or tree trunks (vertical substrates). In 113 recordings of rubbing behavior, the tree part rubbed was identified, and branches represented 59.3% of these records. The difference in the use of these two substrates was not significantly different from equality (binomial test, P40.05), but anogenital rubbing was more asymmetrical, with 36 records on branches and 12 on trunks. Position While Rubbing Rubbing behaviors were most frequently performed while howlers were in sitting postures (51.5%), followed by rubbing behaviors in ventral lying positions (33.3%), upright postures (13.1%) and while hanging by the tail (2%). Body Reactions by Performers The only corporal reactions observed in howlers performing rubbing behaviors were piloerection (11 episodes), sneezing (one episode: sternum rubbing by the adult male in Licurana Group) and rump Rubbing Behavior in Howlers / 579 TABLE IV. Contexts Preceding Rubbing Episodes According to Body Part Rubbed in Groups at Morro Geisler Hyoid Sternum Anogenital Dorsum Abdomen Mandible Total Intergroup conflict Defecation Play Movement in the forest Rain Feeding Grooming Locomotion Vocalization Researcher’s approach No apparent reason 4 1 0 1 0 1 0 0 0 0 1 3 1 0 0 0 1 0 0 0 0 1 0 39 0 1 0 0 0 0 0 2 4 6 5 1 4 6 1 2 2 1 0 7 0 1 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 1 0 13 47 1 6 6 3 2 2 1 4 14 Total 8 6 46 35 3 1 99 TABLE V. Contexts After Rubbing Episodes According to Body Part Rubbed in Groups at Morro Geisler Hyoid Sternum Anogenital Dorsum Abdomen Mandible Total Intergroup conflict Defecation Movement in the forest Locomotion Rain Feeding Grooming Vocalization No apparent reason 3 0 0 1 0 0 0 0 2 3 0 0 1 0 0 0 0 2 0 13 0 3 0 2 0 0 11 6 3 1 5 4 0 2 2 5 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 12 16 1 10 4 2 2 2 21 Total 6 6 29 28 1 0 70 elevation in relation to performer’s forepart (one episode: hyoid rubbing by the adult male in Huebes Group). Piloerection was most frequent in hyoid (four records) and dorsal (three episodes) rubbing episodes. Contexts Contexts preceding and after rubbing behaviors are given, respectively, in Tables IV and V. A large number of rubbing behaviors occurred in undetermined contexts. The relationship between preceding extragroup conflicts with hyoid and sternum rubbing showed significant results (hyoid: Fisher test, Po0.01; sternum: Fisher test, Po0.05), indicating that these forms of rubbing behavior are significantly associated with intergroup encounters when compared with other rubbing modes. Intergroup conflicts accounted for 2% of the time spent with howler groups. The w2 test of adherence showed an overall frequency above the expected of rubbing preceded by intergroup conflict (w2 5 59.2, df 5 2, Po0.001). The same pattern was found for hyoid rubbing alone (w2 5 90, df 5 2, Po0.001) and hyoid plus sternum rubbing (w2 5 157.6, df 5 2, Po0.001). A significant positive relationship also was found in the frequency of anogenital rubbing behavior during both the period preceding and the period after defecation (Fisher test, Po0.001, in both cases). We noted two observations of anogenital rubbing in the Pedreira Group after we had completed data collection for this study. In both episodes, the performer adult female engaged in copulatory behavior with the resident adult male of her group soon after anogenital rubbing. Dorsal rubbing was the only form of rubbing associated with rainfall, being also associated with other contexts. The only possible overmarking episode observed occurred after the study period and involved dorsal rubbing by the adult female and subadult male in Huebes Group along with a recently emigrated adult female. During group progression initiated by the subadult male, he was observed to rub against a tree trunk immediately upon arrival at the tree. He then continued his travel. This rubbing behavior was similarly performed by the resident female and the emigrant female, respectively. The resident female was seen directing agonistic behaviors to the newcomer female, who spent most of the time on group Am. J. Primatol. 580 / Hirano et al. periphery, only coming close to the adult male. The subadult male always had been closely associated with this resident female. Later the emigrant female copulated with the resident male and achieved resident status, staying at normal distances from other group members. Looking Direction In most rubbing episodes the performer’s sight was not clearly directed to any identified target. In six intergroup encounters the adult male looked at the other group or specifically at an adult male in the other group (five rubbing records in Huebes Group and one in Pedreira Group: three with hyoid, three with sternum). Looks directed at birds of prey are responsible for two successive observations in the Huebes Group, with the subadult male rubbing first his anogenital region, and then his dorsum while facing the birds. Responses Responses to rubbing, such as licking and sniffing, were not observed in this study. Besides the above-mentioned overmarking in Group Huebes, we observed responses to scent marks in captive animals at CEPESBI; but as they are individually housed, these were directed to their own marks. Adult males, when rubbing with their region or chin, sometimes salivated, scratched their wet muzzles or sniffed the substrate they had rubbed. DISCUSSION In this study, adult male brown howler monkeys at Morro Geisler were found to be the main performers of rubbing behavior. If rubbing serves a scent-marking function, then our observations are consistent with sexual selection theory that predicts that the sex characterized by a greater level of intrasexual competition marks at a greater frequency [Heymann, 2003a, 2006]. The greater frequency of hyoid and sternum rubbing by males is in agreement with this pattern as these forms of rubbing are considered agonistic signals in howlers [Sekulic & Eisenberg, 1983], and our data suggest this function, with an apparent relationship between this form of rubbing and intergroup encounters. Given the low number of rubbing bouts, short study period and small number of subjects in our study, the function of rubbing in Alouatta still requires additional study. In addition, we found evidence of status-based differences in rubbing behavior. Males are the main performers of rubbing and also the dominant sex, whereas among females, the dominant ones tended to rub more than the subordinates. A clear relationship between social rank and frequency of rubbing has been described in several primate studies [Epple, Am. J. Primatol. 1986; Kleiman & Mack, 1980; Rylands, 1990; Walraven & van Elsacker, 1992]. Heymann [2003a, 2006] concluded that sexual selection is probably the main driving force in the evolution of scent-marking behavior in Neotropical primates, leading to a sexually biased-marking behavior. However, he could not rule out intrasexual competition in both sexes as an important factor, leading to individual variation in marking related to intrasexual hierarchy. We note that most studies of rubbing in platyrrhines (regarded as scent-marking behavior) have involved studies of callitrichids. Howlers, with very distinct ecological and life-history patterns, may show different trends from those seen in callitrichids. Anogenital rubbing is strongly associated with defecation, which points to a cleaning function for this behavior. Braza et al.  verified that A. seniculus rubbed trunks and branches after defecation, leaving a characteristic odor on the substrate that could act as a scent mark. So, a cleaning function does not necessarily preclude a marking one for anogenital rubbing in howler monkeys. Urine can have a marking function [Jones, 2003; Milton, 1975], and it is possible that feces also have this function as they have a strong, characteristic scent in howlers [Emmons & Feer, 1990; Epple, 1986]. Both urine and feces are used as territorial and/or resource defense markers in strepsirrhines [Irwin et al., 2004] and in other mammals [Brashares & Arcese, 1999; Gorman & Trowbridge, 1989; OwenSmith, 1975]. In this regard, anogenital rubbing associated with urination and defecation can spread (non-skin-gland) odoriferous signals onto a substrate. Urine and vaginal secretions also can signal reproductive status of females [Dixson, 1998], and males can directly sniff females’ anogenital area as well as anogenital marks to verify their reproductive condition [Eisenberg, 1976; Neville, 1972]. In Lagothrix lagotricha, female anogenital marking is significantly related to reproductive behavior [Di Fiore et al., 2006]. Two episodes of anogenital rubbing by an adult female in the Pedreira Group were followed by copulations, which is consistent with such a reproductive signaling function. Hyoid and sternum rubbing seem to be associated with intergroup encounters, and could serve both agonistic and territorial functions. Heymann [2000, 2006] found little support for a territorial function in platyrrhine scent-marking behavior, but his analysis grouped distinct forms of marking that could attend very different functions. Other authors have long maintained that there is a territorial function for primate-marking behaviors [Gosling & Roberts, 2001; Mertl-Millhollen, 2006], as seen in other mammals [Gorman, 1990]. Although howler monkey groups do not have exclusive use of their home ranges, they display behaviors that serve to Rubbing Behavior in Howlers / 581 defend the use of an area or of resources (including sexual partners) from neighboring groups, particularly their intense loud calls [Neville et al., 1988; Sekulic, 1982; Whitehead, 1987, 1989]. Hyoid and sternum rubbing also can be used this way, and Sekulic and Eisenberg  observed that hyoid rubbing (‘‘throat rubbing’’) was usually associated with a hostile reaction toward conspecifics in A. seniculus, including intergroup encounters. The presence of piloerection in some episodes of hyoid rubbing also suggests an aggressive motivation by the performer. Young  observed hyoid and chest rubbing by a solitary adult male of A. palliata, peripheral to his study group, after this male copulated with group females. Some time latter he attained group membership, becoming dominant. Our findings reveal that rubbing is probably an important form of marking behavior that can serve several functions depending on the context, form and individual performing this behavior. Our data show some support for the fact that rubbing plays an important role in intersexual and intrasexual dominance interactions, including aggressive and territorial behavior. Functions probably vary according to the body part used. However, the low number of episodes observed makes us cautious about greater generalizations, and all trends viewed here must await confirmation by larger sampling of individuals and increased numbers of observations. The absence of easily to interpret responses by group members or extragroup individuals to rubbing makes it difficult to identify a clear function for this behavior. Neville  also failed to observe direct investigation of rubbed substrates by other howler group members. He concluded that rubbing behaviors were most likely mere scratching, without a communicative function. In our view, some apparent associations of rubbing with particular contexts pointed to communicative functions. Even in callitrichids most marking events do not lead to responses by other individuals [Heymann, 1998]. The single overmarking response, observed in Huebes Group, can only be speculatively interpreted: the dorsal rubbing observed could be a dominance/ agonism display in this context, directed to the emigrant female, who in turn responded in the same manner. Further studies should differentiate among the functions of several kinds of rubbing behaviors. Particular attention must be given to variation among individuals in distinct sex–age classes and with different social statuses. This will serve to clarify the social dynamics involved in rubbing behavior and its relationship with howler monkeys’ social systems. Howlers of both sexes compete for group status and membership [Jones, 1980; Pope, 2000], but the expression of competition is usually not overt, being highly dependent on subtle signals [Jones, 1983; Neville et al., 1988]. Thus, the study of rubbing behavior can offer greater understanding of howler monkey communication. Finally, we need to ask whether howler monkeys’ rubbing behaviors could also be visual signals. Marking behaviors can contain both olfactory and visual signals [Mertl, 1976; Sauther et al., 1999]. A scent-marking function for rubbing behavior does not exclude a visual-marking one, as well as other forms of visual displays. Heymann  examined whether scent marking in tamarins could act as a means of visual signaling by analyzing the proximity of other group members during scent marking. He found that the number of occurrences with two or more individuals being close to the signaler was above the expected. As these behaviors in howlers can be clearly observed by the human researchers from several meters away, their execution can surely act as visual displays on their own, or drive attention to the odor signals left. The mentioned absence of clear olfactory responses to surfaces rubbed by howlers may indicate that the visual components of this behavior are more relevant than the olfactory. Pigment releasing glands are present on brown howlers’ skin [Hirano et al., 2003]. These glands were identified in regions of the throat, groin and lower jaw of adult males, and had coiled tubular secretory portions adjacent to the hair follicles in the dermis. Secretory granules were noted, which produced a secretion similar to the pigment released by the animals. These structures are probably differentiated sweat glands, being larger and having more secretory units and apical cells in the secretory epithelium. Similar, but less-developed structures have been found in juvenile and female individuals [Hirano, 2004]. The development of these glands into their differentiated form in adult males is probably a secondary sexual trait stimulated by the action of sexual hormones during puberty. In at least three field episodes (and in several observations in captivity), we observed that rubbing behaviors left definite colored pigment markings on tree trunks. These color patches may guide an individual in locating a spot where another howler has left a scent mark. These findings suggest that lasting/non-instantaneous visual signals can enhance the olfactory communication associated with rubbing behaviors. Judicious field observations and well-designed field and captive experiments with discrete stimuli [Dugmore et al., 1984; Mertl, 1977] can shed light on the presumed functions and mechanisms associated with rubbing behaviors. Substrates marked by distinct body regions and stimuli (estrous non-estrous female urine, dominant subordinate individuals, male female, familiar outgroup individuals) should be used. Visual function and color discrimination can be assessed through colored marks left on substrates of distinct background (from lighter to darker), and combined with and without associated odors, to test for attractiveness of color marks. Am. J. Primatol. 582 / Hirano et al. ACKNOWLEDGMENTS We thank Patricia Izar, Paul Garber and two anonymous reviewers for their helpful comments. Eugene Harris assisted with careful English revision. Our thanks go to all the students and personnel at CEPESBI (Indaial Biological Research Center) who helped us with data gathering at Morro Geisler. We also thank the support from the University of Blumenau (FURB) and from the Municipality of Indaial. This study was supported by grants from CNPq (RD grant to D. A. G. O., process number 305372/2002-5) and FURB. This study is in compliance with animal care regulations and applicable Brazilian laws. REFERENCES Agresti A. 2002. Categorical data analysis. New York: Wiley. 734p. Altmann J. 1974. Observational study of behavior: sampling methods. Behaviour 49:227–267. Altmann S. 1959. Field observations of a howling monkey society. J Mammal 40:317–330. Brashares JS, Arcese P. 1999. Scent marking in a territorial African antelope: II. The economics of marking with faeces. Anim Behav 57:11–17. Braza F, Alvarez F, Azcarate T. 1981. Behaviour of the red howler monkey (Alouatta seniculus) in the llanos of Venezuela. Primates 22:459–473. Brennan PA, Zufall F. 2006. Pheromonal communication in vertebrates. Nature 444:308–315. Di Fiore A, Link A, Stevenson PR. 2006. Scent marking in two Western Amazonian populations of woolly monkeys (Lagothrix lagotricha). Am J Primatol 68:637–649. Dixson AF. 1998. Primate sexuality. Oxford: Oxford University Press. Dugmore SJ, Bailey K, Evans CS. 1984. Discrimination by male ringtailed lemurs between the scent marks of male and female conspecifics. Int J Primatol 5:235–245. Eisenberg JF. 1976. Communication mechanisms and social integration in the black spider monkey, Ateles fusciceps robustus, and related species. Smithson Contrib Zool 213:1–108. Emmons LH, Feer F. 1990. Neotropical rain forest mammals: a field guide. Chicago: The University of Chicago Press. Epple G. 1986. Communication by chemical signals. In: Mitchel G, Erwin J, editors. Comparative primate biology, vol. 2A: behavior, conservation and ecology. New York: Alan R. Liss. p 531–580. Evans CS. 2006. Accessory chemosignaling mechanisms in primates. Am J Primatol 68:525–544. Gorman ML. 1990. Scent marking strategies in mammals. Rev Suisse Zool 97:3–30. Gorman ML, Trowbridge BJ. 1989. The role of odor in the social lives of carnivores. In: Gittleman JL, editor. Carnivore behavior, ecology, and evolution. Ithaca: Cornell University Press. p 57–88. Gosling LM, Roberts SC. 2001. Testing ideas about the function of scent marks in territories from spatial patterns. Anim Behav 62:F7–F10. Hayes RA, Morelli TL, Wright PC. 2004. Anogenital gland secretions of Lemur catta and Propithecus verreauxi coquereli: a preliminary chemical examination. Am J Primatol 63:49–62. Heymann EW. 1998. Sex differences in olfactory communication in a primate, the moustached tamarin, Saguinus mystax (Callitrichinae). Behav Ecol Sociobiol 43:37–45. Am. J. Primatol. Heymann EW. 2000. Spatial patterns of scent marking in wild moustached tamarins, Saguinus mystax: no evidence for a territorial function. Anim Behav 60:723–730. Heymann EW. 2001. Interspecific variation of scent-marking behaviour in wild tamarins, Saguinus mystax and Saguinus fuscicollis. Folia Primatol 72:253–267. Heymann EW. 2003a. Scent marking, paternal care, and sexual selection in callitrichines. In: Jones CB, editor. Sexual selection and reproductive competition in primates: new perspectives and directions. Norman: American Society of Primatologists. p 305–325. Heymann EW. 2003b. Monogamy in New World primates— what can patterns of olfactory communication tell us? In: Reichard U, Boesch C, editors. Monogamy: mating strategies and partnerships in birds, humans and other mammals. Cambridge: Cambridge University Press. p 248–261. Heymann EW. 2006. Scent marking strategies of New World primates. Am J Primatol 68:650–661. Hirano ZMB. 2004. Secrec- ão epidérmica de Alouatta guariba clamitans (Primates: Atelidae). Ph.D. thesis. Ribeirão Preto, Brazil: Universidade de São Paulo. 144p. Hirano ZMB, Tramonte R, Silva AR, Rodrigues RB, Santos WF. 2003. Morphology of epidermal glands responsible for the release of colored secretions in Alouatta guariba clamitans. Lab Primate Newsl 42:4–7. Irwin MT, Samonds KE, Raharison JL, Wright PC. 2004. Lemur latrines: observations of latrine behavior in wild primates and possible ecological significance. J Mammal 85:420–427. Jones CB. 1980. The function of status in the mantled howler monkey, Alouatta palliata Gray: intraspecific competition for group membership in a folivorous Neotropical primate. Primates 21:389–405. Jones CB. 1983. Social organization of captive black howler monkeys (Alouatta caraya): ‘‘social competition’’ and the use of non-damaging behavior. Primates 24:25–39. Jones CB. 2003. Urine-washing behaviors as conditiondependent signals of quality by adult mantled howler monkeys (Alouatta palliata). Lab Primate Newsl 42:12–14. Kleiman DG, Mack DS. 1980. Effects of age, sex, and reproductive status on scent marking frequencies in the golden lion tamarin, Leontopithecus rosalia. Folia Primatol 33:1–14. Mertl AS. 1976. Olfactory and visual cues in social interactions of Lemur catta. Folia Primatol 26:151–161. Mertl AS. 1977. Habituation to territorial scent marks in the field by Lemur catta. Behav Biol 21:500–507. Mertl-Millhollen AS. 2006. Scent marking as resource defense by female Lemur catta. Am J Primatol 68:605–621. Miller KE, Laszlo K, Dietz JM. 2003. The role of scent marking in the social communication of wild golden lion tamarins, Leontopithecus rosalia. Anim Behav 65:795–803. Milton K. 1975. Urine rubbing behavior in the mantled howler monkey Alouatta palliata. Folia Primatol 23:105–112. Neville MK. 1972. Social relations within troops of red howler monkeys (Alouatta seniculus). Folia Primatol 18:47–77. Neville MK, Glander KE, Braza F, Rylands AB. 1988. The howling monkeys, genus Alouatta. In: Mittermeier RA, Rylands AB, Coimbra-Filho A, Fonseca GAB, editors. Ecology and behavior of Neotropical primates, vol. 2. Washington: World Wildlife Fund. p 349–453. Owen-Smith RN. 1975. The social ethology of the white rhinoceros Ceratotherium simum (Burchell 1817). Z Tierpsychol 38:337–384. Pope TR. 2000. The evolution of male philopatry in neotropical monkeys. In: Kappeler PM, editor. Primate males: causes and consequences of variation in group composition. Cambridge: Cambridge University Press. p 219–235. Ralls K. 1971. Mammalian scent-marking. Science 171:443–449. Rubbing Behavior in Howlers / 583 Rylands AB. 1990. Scent marking behavior of wild marmosets, Callithrix humeralifer (Callitrichidae, Primates). In: MacDonald DW, Müller-Schwarze D, Natynczuk SE, editors. Chemical signals in vertebrates, vol. 5. Oxford: Oxford University Press. p 415–429. Sauther ML, Sussman RW, Gould L. 1999. The socioecology of the ringtailed lemur: thirty-five years of research. Evol Anthropol 8:120–132. Sekulic R. 1982. The function of howling in red howler monkeys (Alouatta seniculus). Behaviour 81:38–54. Sekulic R, Eisenberg JF. 1983. Throat rubbing in red howler monkey (Alouatta seniculus). In: Müller-Schwarze D, Silverstein RM, editors. Chemical signals in vertebrates, vol. 3. New York: Plenum Press. p 347–350. Setz E, Gaspar DA. 1997. Scent-marking behaviour in freeranging golden-faced saki monkeys, Pithecia pithecia chrysocephala: sex differences and context. J Zool 241:603–611. Shoemaker AH. 1979. Reproduction and development of the black howler monkey Alouatta caraya at Columbia Zoo. Int Zoo Yearb 19:150–155. Smith T. 2006. Individual olfactory signatures in common marmosets (Callithrix jacchus). Am J Primatol 68: 585–604. Walraven V, van Elsacker L. 1992. Scent-marking in New World primates: a functional review. Acta Zool Pathol Antverp 82:51–59. Webb DM, Cortes-Ortiz L, Zhang J. 2004. Genetic evidence for the coexistence of pheromone perception and full trichromatic vision in howler monkeys. Mol Biol Evol 21:697–704. Whitehead JM. 1987. Vocally mediated reciprocity between neighbouring groups of mantled howling monkeys, Alouatta palliata palliata. Anim Behav 35: 1615–1627. Whitehead JM. 1989. The effect of the location of a simulated intruder on responses to long-distance vocalizations of mantled howling monkeys, Alouatta palliata palliata. Behaviour 108:73–103. Young PO. 1982. Tree-rubbing behaviour of a solitary male howler monkey. Primates 23:303–306. Am. J. Primatol.