Scent marking in two western Amazonian populations of woolly monkeys (Lagothrix lagotricha)

American Journal of Primatology 68:637–649 (2006)

RESEARCH ARTICLE

Scent Marking in Two Western Amazonian Populationsof Woolly Monkeys (Lagothrix lagotricha)

ANTHONY DI FIORE1,2�, ANDRES LINK2, AND PABLO R. STEVENSON3

1Center for the Study of Human Origins, Department of Anthropology, New YorkUniversity, New York, New York2New York Consortium in Evolutionary Primatology, New York, New York3Departamento de Ciencias Biologicas, Universidad de Los Andes, Bogota, Colombia

We describe patterns of scent marking observed in two wild populationsof lowland woolly monkeys that were the subjects of long-term studies inthe westernmost portion of the Amazon basin. The woolly monkeysengaged primarily in two types of scent marking: chest rubbing andanogenital rubbing. In both study populations, males and femalesperformed both types of scent marking, but males chest-rubbed morecommonly than females, while females engaged in more anogenitalrubbing. We evaluated two nonexclusive hypotheses for the functionof scent marking by woolly monkeys: 1) that scent marking is used insociosexual contexts, and 2) that scent marking is used to conveyinformation about occupancy of or willingness to defend an area fromconspecifics in other social groups. We found no association between theoccurrence of scent-marking behavior and location within the homerange, but did find that scent marking occurred more commonly thanexpected on days when copulations, mating solicitations, and intergroupencounters were observed. Additionally, mating activity and chestrubbing were highly correlated across the yearly cycle, even when thepotentially confounding variable of ripe fruit availability was controlledfor. In woolly monkeys, overt male–male competition is rare and femalechoice is an important part of the mating system. Our results are mostconsistent with the idea that scent marking plays a role in advertisingmale quality or competitive ability, and perhaps in coordinating matingactivity. Am. J. Primatol. 68:637–649, 2006. �c 2006 Wiley-Liss, Inc.

Key words: woolly monkeys; Lagothrix; olfactory signaling; reproduc-tive competition; sexual selection; range advertisement

Published online in Wiley InterScience (www.interscience.wiley.com).DOI 10.1002/ajp.20257

Received 4 November 2004; revised 8 February 2005; revision accepted 9 February 2005

Contract grant sponsor: National Science Foundation; Contract grant sponsor: Wenner-GrenFoundation for Anthropological Research; Contract grant sponsor: L.S.B. Leakey Foundation;Contract grant sponsor: University of California–Davis; Contract grant sponsor: New YorkUniversity.

�Correspondence to: Anthony Di Fiore, Department of Anthropology, New York University,25 Waverly Place, New York, NY 10003. E-mail: [email protected]

rr 2006 Wiley-Liss, Inc.

INTRODUCTION

Mammals commonly use olfactory signaling (particularly scent marking) forcommunication both within and between social groups [Brown, 1979; Brown &MacDonald, 1985; Eisenberg & Kleiman, 1972; Vandenbergh, 1983]. Scent markscan convey information about individual identity, group membership, sex, rank,or reproductive state [Brown & MacDonald, 1985; Halpin, 1986; Heymann, 1998;Smith et al., 1997, 2001; Washabaugh & Snowdon, 1998], and depending on thevolatility of the active compounds in these marks (as well as the nature of thecarrier matrix and the substrates upon which they are deposited), scent marksmay be an effective means of conveying this information over a relatively longtime period, even in the absence of the signaler.

A number of nonexclusive hypotheses concerning the functions of scentmarking in primates have been proposed, most of which fall into three basiccategories: First, scent marking has been proposed to function in a varietyof sociosexual and reproductive contexts, both within and between groups, e.g., toadvertise a signaler’s sexual status or receptivity to members of the opposite sex[Heymann, 1998; Milton, 1975; Lazaro-Perea et al., 1999], to advertise orreinforce a signaler’s dominance over same-sexed conspecifics [Miller et al., 2003;Lazaro-Perea et al., 1999], to compete with same-sexed individuals overreproductive opportunities or in attracting members of the opposite sex[Heymann, 1998, 2003; Ziegler et al., 1993], or to physiologically suppress thereproduction of same-sexed individuals [Abbott et al., 1993; French, 1997].Second, scent marking may function to advertise occupancy, use, or ownership ofan area, and has thus been proposed as a potentially efficient means of defendinga territory or home range [Gorman, 1984; Gosling, 1982, 1986; Moynihan, 1969;Schilling, 1980]. Finally, it was recently suggested that scent marking mayfunction in intragroup communication of ecological information, e.g., to aidorientation of group members within the home range [Milton, 1975] or toadvertise the locations of food sources to group mates [Miller et al., 2003].

Thus far, most work on scent marking in anthropoid primates has focusedon the callitrichines. This behavior constitutes a rich part of their behavioralrepertoire, and callitrichine scent marking has been argued to serve several of thefunctions discussed above in different ecological and social contexts. For example,golden lion tamarins (Leontopithecus rosalia) scent mark more commonly onfeeding trees than on nonfeeding trees, and marks are more frequently depositedby dominant than subordinate males, which suggests that these signals functionboth to communicate the location of important food sources to group mates andto advertise or reinforce male social status [Miller et al., 2003]. In several speciesof Amazonian tamarins (Saguinus spp.) and common marmosets (Callithrixjacchus), pheromones deposited in the scent marks of dominant females appearto be important cues that prompt physiological suppression of reproductivecycling in subordinates [Abbott et al., 1993; French, 1997]. Interestingly, incallitrichines the evidence that scent marking functions in territory defenseis equivocal: some studies have reported that scent marking is more common inperipheral areas or in regions of home-range overlap than in interior portions ofthe range or areas of exclusive use [Rylands, 1990], while others have found nodifference in scent-marking rates in these various contexts [Heymann, 2000;Lazaro-Perea et al., 1999; Miller et al., 2003].

Outside of the callitrichines, few attempts have been made to investigate thefunctions of scent marking behavior in anthropoids. This is true even for manyof the remaining New World monkeys, despite the fact that many of them possess

638 / Di Fiore et al.

Am. J. Primatol. DOI 10.1002/ajp

well developed skin glands on the sternum and perineum and routinely engage ina variety of scent-marking behaviors [Hill, 1965]. Scent marking appears to berelatively common within the large-bodied atelin clade of platyrrhines, althoughthere have been few explicit studies of the functions of marking behavior. Inspider monkeys (Ateles spp.), for example, two related types of scent markingbehaviors–chest-to-mouth scratching and chest rubbing–have been documentedin a captive colony [Klein & Klein, 1971] as well as in one free-ranging populationin Colombia [Klein, 1972]. Additionally, Campbell [2000] noted that Atelesgeoffroyi in Panama rub the fur in their sternal and axillary areas with the leavesfrom a tree of the Family Rutaceae and then subsequently engage in chestrubbing against a substrate. In muriquis (Brachyteles spp.), individuals of bothsexes will deliberately anoint their hands, feet, or tail with urine, which Milton[1985] speculated may function to advertise a female’s reproductive state and/orto play a role in mate competition among males during mating aggregations. Inthis paper we describe two types of scent marking (chest rubbing and anogenitalrubbing) that are commonly seen in both captive and wild populations of anotheratelin, the lowland woolly monkey (Lagothrix lagotricha Platyrrhini: Atelidae),and we test several hypotheses about the possible social functions of thesebehaviors.

Chest rubbing has been well described for woolly monkeys in captivity[Williams, 1968, White et al., 2000] and has been recorded in multiple wildpopulations of woolly monkeys (e.g., Lagothrix lagotricha lagotricha, in additionto the two subspecies described here [Defler, 2003]). A chest-rubbing bouttypically involves wetting a substrate (e.g., branch, liana, or tree trunk) withsaliva followed by repeated upward thrusts of the chest against the wet substrate.Alternating bouts of licking and chest rubbing may continue for up to severalminutes. Klein and Klein [1971] observed a similar administration of saliva toa substrate in the context of intergroup ‘‘threatening’’ by male spider monkeys,which they suggested may function to effectively spread the mark over a largerarea. Bouts of anogenital rubbing are much shorter in duration than chest-rubbing bouts and involve the forward and backward movement of the anogenitalregion against a substrate for a couple of seconds only.

Using data collected during long-term studies of the socioecology of woollymonkeys at two western Amazonian sites, we first examine the potential sociosexualrole of scent marking by evaluating whether marking is concentrated on days ofconspicuous female receptivity or observed sexual behavior, as we predict it would beif the behavior plays a role in advertising or coordinating sexual activity or is used incompetition among same-sex animals over mating opportunities or over beingselected as a mating partner. Second, we evaluate the potential role of scent markingin intergroup communication and test several predictions concerning an associationbetween scent marking and patterns of range use. Specifically, we predict that ifscent marking is associated with either home-range advertisement or territorialdefense, then marking should more commonly be performed in peripheral vs.interior portions of the home range, in areas of overlap with adjacent groups vs. areasof range overlap, and on days when intergroup encounters were recorded.

MATERIALS AND METHODS

Study Animals

Population 1: Lagothrix lagotricha poeppigii, Yasunı National Park, EcuadorBehavioral data were collected on three groups of woolly monkeys at the

Proyecto Primates Research Site in Yasunı National Park, Ecuador, between

Scent Marking in Woolly Monkeys / 639


April 1995 and March 1996 using a combination of scan and focal animalsampling on alternate observation days, and again between August 2002 and July2003 using only focal sampling methods [Di Fiore, 1997, 2003; Di Fiore &Rodman, 2001]. During scan sampling, the behaviors of all visible groupmembers, including the performance of chest rubbing, anogenital rubbing,mating, and mating solicitations, were noted at 10-min intervals throughout theday. Focal sampling focused on four age/sex classes of woolly monkeys: adultmales, subadult males, adult females with dependent offspring, and femaleswithout dependents. Focal samples consisted of continuous observations duringwhich we recorded instantaneous data at 1-min intervals on the behavioral state(forage, rest, move, social, or other) of the focal animal, and noted any additionalsocial interactions and scent-marking behavior that occurred between theinstantaneous sampling points. During the first study period, focal sampleslasted between 1 min and 30 min, depending on the visibility of the focal animal,while only 5-min samples were collected during the latter study period. We werenot able to consistently recognize animals individually during either study period,but we tried to ensure as broad a sample as possible of individuals within eachage-sex class, including animals that we could and could not recognize. Thelocation of each scent-marking bout within the home range was reconstructedfrom ranging data collected either every 10 min (scan sample days), at the end ofeach focal sample (1995–1996 focal sampling days), or every 15 min (2002–2003focal sampling days) throughout the day.

Population 2: Lagothrix lagotricha lugens, Tinigua National Park, ColombiaIn Tinigua, behavioral data were collected on four groups of woolly monkeys

during three yearly cycles (April 1990–March 1991, August 1996–July 1997, andFebruary 2000–January 2001) using focal animal sampling [Stevenson et al.,1994]. However, about 90% of the observations were concentrated on one mainstudy group (CR-12). Concurrent with focal sampling, all performances of chestrubbing, mating, and mating solicitations by any visible group member wererecorded ad libitum. Although anogenital rubbing was occasionally observed,especially in mating contexts, no attempt was made at Tinigua to quantify itsfrequency. The location of each scent-marking bout within the home range wasreconstructed from ranging data collected every 30 min.

During the first yearly cycle the sampling time was equally divided amongfour age/sex classes–adult males, adult females with dependent offspring (e.g.,carrying infants), other adult females, and juveniles of both sexes–and each age/sex class was observed for 12 hr per month. During the second and third yearlycycles, the same basic protocol was followed, but an additional 12 hr of focal datawere collected each month, focusing mainly on adult males and adult females withdependent offspring. Focal samples lasted 0.5–24 hr (spread over contiguous days)and were conducted only on individually recognizable animals.

Concurrently with behavioral sampling, data were collected at biweeklyintervals on the habitat-wide availability of ripe fruits, the predominantcomponent in the diet of woolly monkeys [Di Fiore, 2004; Peres, 1994; Stevensonet al., 1994]. Based on a set of 12 phenological transects totaling 5.6 km in length,a monthly estimate of community-wide fruit production was calculated using thenumber of trees (45 cm diameter at breast height (DBH)) and lianas (43 cmDBH) that were producing ripe fruits on the transects. This index has been wellcorrelated with several other indices of fruit abundance, including ones based



on fruit traps, basal area, and estimates of actual fruit production [Stevenson,2002, 2004].

Analytical Methods

To estimate the frequency of scent marking among different age-sex classesand evaluate the temporal association between scent marking and reproductivebehavior, we used only data from focal samples, because many scent-markingbouts, which can be short in duration, were undoubtedly missed during scansampling. Thus, for all observation days during which at least 60 min of focalanimal data were collected (Yasunı, n 5 101 days (the actual time with theanimals was typically much longer than 60 min, given the difficulty ofmaintaining visual contact with individual animals long enough to completesamples); Tinigua: n 5 168 days), we scored the presence or absence and numberof times any animal followed was seen chest rubbing, anogenital rubbing,soliciting mating, or copulating. For Tinigua, we also scored whether anyintergroup encounters or interspecific aggressive behaviors directed toward otherprimates (including the observer) were observed on each observation day.However, too few of these events occurred at Yasunı to permit adequate statisticalanalysis for that population.

For our spatial analyses we used all incidents of scent marking recordedduring either focal or scan samples and scored for each whether it occurred in aborder vs. interior portion of the home range, and, for Yasunı, whether it occurredin an area of range overlap with an adjacent group vs. an area of exclusive use.The latter analysis could not be performed with data from Tinigua because thehome range of the primary study group overlapped 100% with those of adjacentgroups. To define border/interior and overlap/exclusive areas, we divided thehome range of each group into a series of 100 m� 100 m quadrats. Each group’shome range was considered to be the total set of quadrats in which the group wasrecorded during group follows. Border areas were then identified, for Yasunı, asthe entire set of 1-ha quadrats comprising the periphery of the group’s homerange, while interior quadrats were those that did not constitute at least part ofthe range periphery. For Tinigua, a 2-ha border area was used. These alternativecriteria were applied so that, in each site, border areas comprised roughly thesame total proportion of each study group’s home range (Yasunı: 52–54%, n 5 3groups; Tinigua: 41%, n 5 1 group). Overlap areas were defined as quadratsincluded within the ranges of more than one group.

For our analyses we pooled all incidents of scent marking across groups andacross years within each of the study populations because there were too few casesper group to analyze them separately. Marking bouts by different individuals (orby the same individual but separated by more than 15 min) were considered to beindependent for the purposes of calculating rates of marking behavior and forconsidering the spatial distribution of marks, where the expected number ofmarking bouts in border vs. interior and overlap vs. nonoverlap areas was basedon the frequency of use of these different regions determined from ranging data.For analysis of the temporal association between marking and reproductivebehaviors, our primary unit of analysis was either the day (i.e., whether markingwas observed or not during an observation day) or, for cases of focal observationsspanning 2 contiguous days on a single individual at Tinigua, the follow itself (i.e.,whether marking was observed during the follow period). Associations betweenscent marking and reproductive behavior, and between scent marking and spatialposition, were evaluated using nonparametric G-tests with William’s correction



[Sokal & Rohlf, 1981]. Differences between the sexes in marking frequency wereevaluated via Kruskal-Wallis nonparametric analysis of variance (ANOVA)[Siegel & Castellan, 1998], using data for the Tinigua population only. For thisanalysis we first calculated average yearly marking frequencies for each age-sexclass for each year of the study and then used these yearly means as independentcell values. Spearman’s rank correlation was used to assess the associationbetween the use of space within the home range and scent-marking locations forboth study populations. Finally, for the Tinigua population, we examinedtemporal variation in scent marking and reproductive behavior, intergroup andinterspecific aggression, and ripe fruit availability via Pearson’s bivariate andpartial correlation analysis as implemented in SPSS v7.5 (1997). In order to beconservative, all statistical tests employed were two-tailed even when wepredicted the direction of a particular relationship beforehand.

RESULTS

Frequency and Patterning of Scent Marking

We recorded a total of 63 chest rubbing bouts and 112 anogenital rubbingbouts during 262 total hr of focal observation and 690 hr of scan sampling over asingle year in Yasunı, and 103 chest rubbing bouts in 2,016 total hr of focalobservation across 3 years at Tinigua. Both chest rubbing and anogenital rubbingwere performed almost exclusively by nonjuveniles (496% of cases in both studypopulations). Chest rubbing was much more frequently performed by males thanfemales in both Yasunı and Tinigua (54/63 and 89/98 cases, respectively, when thesex of the performer could be determined), mirroring a pattern also seen incaptivity. Chest-rubbing rates differed significantly by nonjuvenile age/sex classat Tinigua (Kruskal-Wallis test: H 5 6.01, df 5 2, P o 0.05; Fig. 1), with post-hoctests revealing that males marked significantly more than females with orwithout dependents. The fact that we could not consistently recognize animals

Fig. 1. Rate of chest rubbing, measured as the number of bouts per 12-hr follow, by variousnonjuvenile age/sex classes of woolly monkeys in Tinigua. Adult and subadult males performedchest rubbing at a much higher rate compared to nonjuvenile females with or without infants.Statistically significant differences in post-hoc comparisons are indicated with asterisks.



individually in Yasunı precludes a statistical evaluation of whether males andfemales chest-rubbed at different rates in that population. Nonetheless, multipleindividuals contributed to both the male and female data for Yasunı, and we areconfident that any bias in sampling intensity among potential focal animals wasminor; thus, the nearly sevenfold difference seen in the frequency of chestrubbing by males vs. females in that population (Table I) is likely to be anaccurate estimate. By contrast, females in Yasunı performed anogenital rubbingmuch more frequently than males (75/107 cases) and at roughly twice the malerate (Table I).

Scent Marking and Reproductive Behavior

In Yasunı, either copulations or obvious mating solicitations (e.g., femalepresented to male and/or made a repeated open-mouthed display with teethchattering or lip smacking [Di Fiore, 1997; Ramirez, 1988] or both were seen on24 of the days during which at least 60 min of focal data were collected. Chestrubbing (by either males or both males and females) was observed on 12 of thesesame days and anogenital rubbing on 16 (Table II). Both chest rubbing andanogenital rubbing occurred significantly more commonly than expected on daysof reproductive activity (chest rubbing: Gadj 5 6.36, df 5 1, P o 0.05; anogenitalrubbing: Gadj 5 5.06, df 5 1, P o 0.05). Moreover, when male and female scentmarking were analyzed separately, the association between chest rubbing andreproductive behavior was still seen for males (Gadj 5 7.07, df 5 1, P o 0.01) butnot for females (Gadj 5 0.27, df 5 1, NS). In addition, the association betweenanogenital rubbing and reproductive behavior was significant for females (femaleGadj 5 4.43; df 5 1, P o 0.05) and very closely approached significance for males(male Gadj 5 3.80, df 5 1, P 5 0.051). Finally, the rate of anogenital rubbing washigher on days of conspicuous reproductive activity compared to days without forboth females (0.7370.16 SE vs. 0.3970.07 SE bouts/hr) and males (0.3770.12 SEvs. 0.1770.05 SE bouts/hr), although the nature of our sampling regimeprecluded a statistical evaluation of these differences.

In the Tinigua population, chest rubbing was observed on 55 days, and on35 of these days reproductive behavior was also observed (Table II)–an associationthat, as in Yasunı, was highly significant (Gadj 5 36.07, df 5 1, P o 0.001).

TABLE I. Rates of Scent Marking Recorded During Focal Sampling for VariousAge-Sex Classes of Woolly Monkeys in Yasunı�

Age-sex class Chest rubbing Anogenital rubbing

Adult male 0.317 0.262Subadult male 0.216 0.189

Males (grouped) 0.288 0.241Adult female with dependent 0.035 0.352Unburdened adult or subadult female 0.051 0.650

Females (grouped) 0.044 0.525

�Rates were calculated as the total number of bouts observed per total number of hours of focal animal datacollected on the given age-sex class, thus precluding an estimate of the error or variance in mean rates.



Scent Marking, Range Use, and Intergroupand Interspecific Interactions

Contrary to our predictions about how scent marking might be related torange use, we found that the overall performance of chest rubbing plus anogenitalrubbing in Yasunı was not more common in border areas than in interior portionsof the range (Gadj 5 0.33, df 5 1, NS), or in areas of overlap vs. areas of exclusiveuse (Gadj 5 3.41, df 5 1, NS). In fact, with respect to the latter result, the trendwas in the opposite direction from what we predicted (i.e., a greater proportion ofscent-marking bouts took place in exclusively used areas than in overlap areasrelative to the amount of time spent in each of those portions of the range).Likewise in Tinigua, chest rubbing was not more common than expectedin border portions of the range (Gadj 5 0.01, df 5 1, NS). Overall, for both sitesthe number of scent marks recorded in a quadrat was significantly correlatedwith the relative use of that quadrat (Spearman’s rank correlation between thenumber of days a quadrat was used and the total number of scent marks recordedin the quadrat: Yasunı: rs 5 0.378, n 5 226 quadrats, P o 0.01; Tinigua:rs 5 0.357, n 5 181 quadrats, P o 0.01).

In Tinigua, chest rubbing occurred more commonly than expected on daysduring which the focal group had an aggressive encounter with another group ofwoolly monkeys (Gadj 5 5.01, df 5 1, P o 0.05), as well as on days during whichfocal subjects directed aggressive displays toward the observer (Gadj 5 9.48, df 5 1,P o 0.01). Chest rubbing was also more common than expected during thoseobservation periods when both copulations and intergroup encounters were seenvs. those periods when intergroup encounters occurred but no mating was noted(Gadj 5 7.45, df 5 1, P o 0.05). We found no association between days with chestrubbing and days when the woolly monkeys had aggressive interactions withother nonhuman primate species (Gadj 5 0.05, df 5 1, NS).

TABLE II. Association Between Scent Marking and Reproductive Activity forWoolly Monkeys in Yasunı and Tinigua�

Copulations or Solicits

Yes No Total

YasunıChest Rubbinga

Yes 12 17 29No 12 60 72Total 24 77 101

Anogenital Rubbingb

Yes 16 31 47No 8 46 54Total 24 77 101

TiniguaChest Rubbingc

Yes 35 20 55No 19 94 113Total 54 114 168

�Numbers in each cell are observation days during which the given combination of scent marking andreproductive behavior was seen.aGadj 5 6.36, df 5 1, Po0.05.bGadj 5 5.06, df 5 1, Po0.05.cGadj 5 36.07, df 5 1, Po0.0001.



Temporal Variation in Scent-Marking Behavior

Because data were collected across multiple annual cycles in Tinigua vs.across a much smaller temporal range in Yasunı, we used only data from Tiniguato examine seasonal variation in scent-marking behavior. At Tinigua, themonthly rate of chest rubbing was closely correlated with both monthly matingfrequency and the monthly rate of aggressive intergroup encounters, and all threeof these variables were significantly correlated with the availability of ripe fruitresources that month (Table III, Fig. 2). However, when the effects of fruitavailability are controlled for through partial correlation, the association betweenmating and chest rubbing remains highly significant (r 5 0.744, n 5 12, df 5 9,P o 0.01), while that between aggression and chest rubbing disappears entirely(r 5 0.039, n 5 12, df 5 9, P 5 0.91).

DISCUSSION

In two populations of western Amazonian woolly monkeys, chest rubbingby males occurred more commonly than expected on days of conspicuous

TABLE III. Pearson Bivariate Correlation Matrix Among the Monthly Frequencyof Chest-Rubbing, Reproductive Behavior, Intergroup Aggression, and a Measureof the Habitat-Wide Monthly Availability of Ripe Fruit at Tinigua�

Chest rubbing CopulationsIntergroupaggression

Fruitavailability

Chest rubbing 1.000 – – –Copulations 0.857b 1.000 – –Intergroup aggression 0.619a 0.505 1.000 –Fruit availability 0.735b 0.638a 0.821b 1.000

�For all cells, N 5 12 months.aPo0.05.bPo0.01.

Fig. 2. Monthly rates of chest rubbing and mating, measured as the total number of bouts observedduring 60 standard hours of focal observation per month, and the monthly index of ripe fruitavailability [Stevenson, 2002] for woolly monkeys in Tinigua.



reproductive activity (either actual copulations or insistent mating solicitationson the part of females), and in one of these populations, anogenital rubbing byboth males and females was also more commonly seen on these days (and mayhave occurred at a higher rate) than on days when no reproductive activity wasseen. Moreover, in one population studied over several years, there was a clearand strong association between monthly rates of chest rubbing and monthlycopulation frequency, even when the availability of ripe fruits in the habitat wascontrolled for. These results thus provide solid support for the idea thatscent marking by both male and female woolly monkeys functions in a sociosexualor reproductive context [Heymann, 1998, 2003; Lazaro-Perea et al., 1999;Milton, 1975].

Whether these olfactory signals are targeted principally toward same- oropposite-sexed individuals requires further study, but we suggest that chestrubbing, at least, may play an important role in reproductive competition amongmales. For several mammalian taxa, scent marking by males is argued to providean honest signal of competitive ability, both to other males and to potential mates[Hurst & Rich, 1999; Rich & Hurst, 1999]. Theoretically, the best competitorswould be those that are most successful at either inhibiting marking by otheranimals (ensuring that their own marks are the most common in an area[Gosling, 1982]), or at responding to competing scent marks by depositing theirown on top of those of challengers (e.g., the ‘‘competitive countermarks’’hypothesis [Hurst & Rich, 1999]. Other animals could thus reliably assess malecompetitive ability indirectly by evaluating the extent to which scent marking inan area is monopolized by particular males. Alternatively, it is also plausible thatmale chest rubbing represents a ‘‘costly’’ signal that is correlated with malecondition [Grafen, 1990; Johnstone, 1995; Zahavi, 1975]. Either way, male scentmarks may provide females with a metric for choosing among potential mates,and female mate choice has been suggested to be an important component of themating system for all ateline primates, including woolly monkeys [Di Fiore, 1997;Di Fiore & Campbell, 2006; Strier, 1990]. Consistent with the competitivecountermarks hypothesis, we noted in both of our study populations severalinstances in which two males, from the same or different groups, chest-rubbed inexactly the same location on the same day. Unfortunately, no data are available asto whether females prefer to mate with countermarking males or with males whochest-rub more frequently, although these would be clear predictions of thehypothesis that male scent marking functions to advertise competitive ability orcondition. More detailed data on the chest-rubbing and copulation frequenciesof individual males are clearly needed to better evaluate this possibility.

Why anogenital marking in the Yasunı population was also associated withdays of reproductive activity remains open to speculation, but in the closely-related muriquis, females are thought to use urine washing to advertise theirsexual receptivity to males [Milton, 1985], and we might speculate that anogenitalrubbing serves a similar function in woolly monkeys. Unfortunately, given thedifficulty of recognizing individual female animals in Yasunı, the data availabledo not allow us to even say whether it was the same females we saw scent markingwho were seen mating on a given day. Clearly, to tease out in better detail justwhat reproductive significance anogenital scent marking might have for each sexwill require additional study, particularly of populations in which individuals canbe recognized consistently and unequivocally.

With respect to the hypothesis that scent marking is associated with rangeadvertisement or defense [Gosling, 1982, 1986; Gorman, 1984; Moynihan, 1969;Schilling, 1980], we found no relationship between the locations of scent marks



and either the border area of a group’s range (for both the Yasunı and Tiniguapopulations) or the region of overlap between adjacent groups (only testable forthe Yasunı population). Indeed, the finding that the number of scent marksrecorded in each quadrat was significantly associated with the intensity ofquadrat use in both study populations argues against a range advertisement/defense function for scent marking, as the most heavily used portions of a group’srange did not neatly fall into any particular quadrat classification (e.g., border,interior, overlap, or exclusive-use quadrat).

For the Tinigua population, we also found a temporal association betweenchest rubbing and the occurrence of intergroup encounters on both a daily andmonthly basis. However, the latter association disappeared entirely when ripefruit availability was controlled for through partial correlation, while a stronglypositive association between chest rubbing and mating remained. Probably,various kinds of social interactions (including both mating and intergroupencounters) occur more commonly during times of ripe fruit abundance,producing a spurious correlation between the monthly rates of chest rubbingand intergroup encounters. We interpret the association between chest rubbingand intergroup encounters at the daily level as being compatible with the ideathat chest rubbing is used in sociosexual contexts, as a form of reproductivecompetition among males in different groups. In fact, this interpretation issupported by our finding that chest rubbing was seen more often than expectedduring periods when both copulations and intergroup encounters occurred vs.those periods when intergroup encounters occurred but no mating was seen.Female transfer between groups has been frequently observed in the Tiniguapopulation [Nishimura, 2003] (Stevenson, unpublished data), and molecular datafrom Yasunı confirm that gene flow between adjacent woolly monkey groups isextensive [Di Fiore & Fleischer, 2005]. These observations may reflect theoutcome of female choice regarding which males to associate with, and suggesta high potential for reproductive competition among males of different groups.

A number of additional hypotheses for the function of scent marking havebeen proposed for other platyrrhine taxa and cannot be ruled out in the presentstudy. For example, Milton [1975] suggested that ‘‘urine rubbing’’–a behavior inwhich urine is deliberately applied to the soles of the feet and/or the bottomsurface of the tail–may be used by mantled howler monkeys to demarcate ‘‘scenttrails’’ that could function to orient group members within their home range orhelp maintain group unity during travel. While this is a fascinating hypothesis,urine rubbing was never seen in our study populations, and both chest rubbingand anogenital rubbing bouts seem to occur far too infrequently and too farapart spatially for either of these to be effective means of marking routes withinthe range.

Our results are consistent with a recent review of scent-marking behavior inNew World primates (Heymann, this issue), which found no evidence that scentmarking serves a territorial function. They are also in accord with Milton’s [1985]suggestion that another type of atelin scent-marking behavior–urine washing bymuriquis–functions primarily in a reproductive or sociosexual context. Lastly, ourresults from two wild populations of woolly monkey are very similar to thoseobtained in a captive population of 10 animals housed at the Louisville ZoologicalGarden. In the latter population, chest rubbing was predominantly performed bythe single adult male in the group, and occurred predominantly when females inthe group were exhibiting behavioral signs of sexual receptivity [White et al.,2000]. In that study, as in ours, the authors attributed a reproductive competitionfunction to chest rubbing by the male.



ACKNOWLEDGMENTS

A.D. and A.L. are grateful to the Ministerio de Ambiente of Ecuador for itscontinued support of primate research in Yasunı National Park. Importantlogistical assistance during the study was provided by the staff of the EstacionCientıfica Yasunı of the Pontificia Universidad Catolica del Ecuador, and byRepsol-YPF, Ecuador, and was made possible by the cooperation of the Huaoranicommunities of Guiyero and Timpoca. P.R.S. acknowledges the support of theCentro de Investigaciones Ecologicas La Macarena (CIEM) and the Japan-Colombia cooperative program for the study of primates. Studies at Tinigua weremade possible by research permits granted by the Unidad de Parques, Ministeriodel Medioambiente of Colombia. In addition, P.R.S. thanks all of the agencies thathave funded projects at CIEM, including the Universidad de Los Andes, theFundacion para la Promocion de la Investigacion y la Tecnologıa (Banco de laRepublica), the Margot Marsh Foundation, the Lincoln Park Zoo, PrimateConservation, Inc., IdeaWild, and Colciencias. Finally, we thank Eckhard Heymannand two anonymous reviewers for helpful comments that improved this manuscript.

References

Abbott DH, Barrett J, George LM. 1993.Comparative aspects of the social suppres-sion of reproduction in female marmosetsand tamarins. In: Rylands AB, editor.Marmosets and tamarins: systematics, be-haviour, and ecology. Oxford: Oxford Uni-versity Press. p 152–163.

Brown RE. 1979. Mammalian social odors: acritical review. Adv Study Behav 10:103–162.

Brown RE, MacDonald DW, editors. 1985.Social odours in mammals. Oxford: Claren-don Press. 882p.

Campbell CJ. 2000. Fur rubbing behavior infree-ranging black-handed spider monkeys(Ateles geoffroyi) in Panama. Am J Primatol51:205–208.

Defler TR. 2003. Primates de Colombia. Bogota:Conservation International. 543p.

Di Fiore A. 1997. Ecology and behavior oflowland woolly monkeys (Lagothrix lagotri-cha poeppigii, Atelinae) in eastern Ecuador.Ph.D. dissertation, University of Califor-nia–Davis, Davis, CA.

Di Fiore A. 2003. Ranging behavior andforaging ecology of lowland woolly monkeys(Lagothrix lagotricha poeppigii) in YasunıNational Park, Ecuador. Am J Primatol 59:47–66.

Di Fiore A. 2004. Diet and feeding ecology ofwoolly monkeys in a western Amazonianrainforest. Int J Primatol 24:767–801.

Di Fiore A, Campbell CJ. 2006. The atelines:variation in ecology, behavior, and socialorganization. In: Campbell CJ, Fuentes A,MacKinnon KC, Panger M, Bearder S,editors. Primates in perspective. Oxford:Oxford University Press. p 155–185.

Di Fiore A, Fleischer RC. Social behavior,reproductive strategies, and population

genetic structure of Lagothrix lagotrichapoeppigii. Int J Primatol (in press).

Di Fiore A, Rodman PS. 2001. Time allocationpatterns of lowland woolly monkeys (Lago-thrix lagotricha poeppigii) in a neotropicalterra firma forest. Int J Primatol 22:449–480.

Eisenberg JF, Kleiman D. 1972. Olfactorycommunication in mammals. Annu RevEcol Syst 3:1–32.

French JA. 1997. Proximate regulation ofsingular breeding in callitrichid primates.In: Solomon NG, French JA, editors. Coop-erative breeding in mammals. Cambridge:Cambridge University Press. p 34–75.

Gorman ML. 1984. Scent marking and terri-toriality. Acta Zool Fennica 171:49–53.

Gosling LM. 1982. A reassessment of thefunction of scent marking in territories.Z Tierpsychol 60:89–118.

Gosling LM. 1986. Economic consequencesof scent marking in mammalian territoriality.In: Duvall D, Muller-Schwarze D, SilversteinRM, editors. Chemical signals in vertebrates4. New York: Plenum Press. p 385–395.

Grafen A. 1990. Biological signals as handi-caps. J Theor Biol 144:517–546.

Halpin ZT. 1986. Individual odors amongmammals: origins and functions. Adv StudyBehav 16:39–70.

Heymann EW. 1998. Sex differences in olfactorycommunication in a primate, the moustachedtamarin, Saguinus mystax (Callitrichinae).Behav Ecol Sociobiol 43:37–45.

Heymann EW. 2000. Spatial patterns of scentmarking in wild moustached tamarins,Saguinus mystax: no evidence of a territor-ial function. Anim Behav 60:723–730.

Heymann EW. 2003. Scent marking, paternalcare, and sexual selection in callitrichines.



In: Jones CB, editor. Sexual selection andreproductive competition in primates: newperspectives and directions. Norman, OK:American Society of Primatologists. p 305–325.

Hill WCO. 1962. Primates: comparative anat-omy and taxonomy. V. Cebidae. Edinburgh:Edinburgh University Publications. 537p.

Hurst JL, Rich TJ. 1999. Scent marks ascompetitive signals of mate quality. In:Johnston RE, Muller-Schwarze D, SorensonP, editors. Advances in chemical commu-nication in vertebrates. New York: PlenumPress. p 209–226.

Johnstone RA. 1995. Sexual selection, honestadvertisement and the handicap principle:reviewing the evidence. Biol Rev Camb PhilSoc 70:1–65.

Klein L, Klein D. 1971. Aspects of socialbehavior in a colony of spider monkeysAteles geoffroyi at the San Francisco Zoo.Int Zoo Yearb 11:175–181.

Klein LL. 1972. The ecology and socialbehavior of the spider monkey, Atelesbelzebuth. Ph.D. dissertation, University ofCalifornia–Berkeley, Berkeley, CA.

Lazaro-Perea C, Snowdon CT, Arruda MF.1999. Scent-marking behavior in wildgroups of common marmosets (Callithrixjacchus). Behav Ecol Sociobiol 46:313–324.

Miller KE, Laszlo K, Dietz JM. 2003. The roleof scent marking in the social communica-tion of wild golden lion tamarins, Leonto-pithecus rosalia. Anim Behav 65:795–803.

Milton K. 1975. Urine rubbing behavior in themantled howler monkey Alouatta palliata.Folia Primatol 23:105–112.

Milton K. 1985. Urine washing behavior inthe woolly spider monkey (Brachytelesarachnoides). Z Tierpsychol 67:154–160.

Moynihan M. 1969. Comparative aspects ofcommunication in New World primates. In:Morris D, editor. Primate ethology. GardenCity, NJ: Doubleday. p 306–342.

Nishimura A. 2003. Reproductive parametersof wild Lagothrix lagotricha. Int J Primatol24:707–722.

Peres CA. 1994. Diet and feeding ecology ofgray woolly monkeys (Lagothrix lagotrichacana) in central Amazonia: comparisons withother atelines. Int J Primatol 15:333–372.

Ramirez M. 1988. The woolly monkeys, genusLagothrix. In: Mittermeier RA, Rylands AB,Coimbra-Filho AF, da Fonseca GAB, edi-tors. Ecology and behavior of neotropicalprimates. Washington, DC: World WildlifeFund. p 539–575.

Rich TJ, Hurst JL. 1999. The competingcountermarks hypothesis: reliable assess-ment of competitive ability by potentialmates. Anim Behav 58:1027–1037.

Rylands AB. 1990. Scent marking behaviorof wild marmosets, Callithrix humeralifer

(Callitrichidae: Primates). In: MacDonaldDW, Muller-Schwarze D, Natynczuk SE,editors. Chemical signals in vertebrates 5.Oxford: Oxford University Press. p 415–429.

Schilling A. 1980. The possible role of urine interritoriality of some nocturnal prosimians.Symposia of the Zoological Society ofLondon 45:165–193.

Siegel S, Castellan NJ. 1988. Nonparametricstatistics for the behavioral sciences. 2nd ed.New York: McGraw-Hill, Inc. 399p.

Smith TE, Abbott DH, Tomlinson AJ, Mlot-kiewicz JA. 1997. Differential display ofinvestigative behavior permits discrimina-tion of scent signatures from familiar andunfamiliar socially dominant female mar-moset monkeys (Callithrix jacchus). J ChemEcol 23:2523–2546.

Smith TE, Tomlinson AJ, Mlotkiewicz JA,Abbott DH. 2001. Female marmoset mon-keys (Callithrix jacchus) can be identifiedfrom the chemical composition of theirscent marks. Chem Senses 26:449–458.

Sokal RR, Rohlf FJ. 1981. Biometry. 2nd ed.San Francisco: W.H. Freeman and Com-pany. 887p.

Stevenson PR, Quinones MJ, Ahumada JA.1994. Ecological strategies of woolly mon-keys (Lagothrix lagotricha) at La Macarena,Colombia. Am J Primatol 32:123–140.

Stevenson PR. 2002. Frugivory and seeddispersal by woolly monkeys at TiniguaNational Park, Colombia. Ph.D. disserta-tion, State University of New York, StonyBrook, NY.

Stevenson PR. 2004. Phenological patterns ofwoody vegetation at Tinigua National Park,Colombia. Caldasia 26:125–150.

Strier KB. 1990. New World primates, newfrontiers: insights from the woolly spidermonkey, or muriqui (Brachyteles ara-chnoides). Int J Primatol 11:7–19.

Vandenbergh JG, editor. 1983. Pheromonesand reproductive behavior in mammals.New York: Academic Press. 298p.

Washabaugh K, Snowdon CT. 1998. Chemicalcommunication of reproductive status infemale cotton-top tamarins (Saguinus oedi-pus oedipus). Am J Primatol 45:337–349.

White BC, Dew SE, Prather JR, Stearns M,Schneider E, Taylor S. 2000. Chest-rubbingin captive woolly monkeys (Lagothrix lago-tricha). Primates 41:185–188.

Williams L. 1968. Man and monkey. Philadel-phia: J.B. Lippincott Company. 203p.

Zahavi A. 1975. Mate selection: a selection forhandicap. J Theor Biol 53:205–214.

Ziegler TE, Epple G, Snowdon CT, Porter TA,Belcher AM, Kuederling I. 1993. Detectionof the chemical signals of ovulation in thecotton-top tamarin, Saguinus oedipus.Anim Behav 45:313–322.



Date post:	15-May-2023
Category:	Documents
Upload:	independent
View:	0 times
Download:	0 times

Scent marking in two western Amazonian populations of woolly monkeys (Lagothrix lagotricha)

Documents