Behavioral Ecology - The transmission and function of chemical signals in Lemur catta lyrics

Abstract The goals of this study were to investigate the transmission and possible functions of chemical signals in intragroup communication among ring-tailed lemurs, Lemur catta. In particular, I examined the effects of s** on these processes because s**ual selection theory predicts specific functions for chemical signals. I recorded all interactions with 214 scent marks of 11 male and 9 female ring-tailed lemurs during the first 10 min following their deposition. I found that 62% of these scent marks were investigated with a median latency of 30 s and that 89% of investigated scents were also counter-marked by the receiver. The type of focal scent mark (male or female anogenital and male antebrachial mark) had a significant effect on both the timing and type of response. Males investigated and counter-marked female scents more often than vice versa, but significant second-order transitions suggested that the behavior of an animal was not only influenced by the immediately preceding scent mark and that a scent is not completely masked by a counter-mark. There was no evidence for an audience effect, and only social rank of female senders had an effect on receivers. Variation in the response of receivers across reproductive seasons as a function of senders' s** indicated that female scents may function in mate attraction and competition among females, whereas male scents may be primarily used in intras**ual competition. Three main conclusions emerged. First, the exchange of olfactory signals within groups was highly structured and surprisingly efficient. Second, olfactory signals may constitute general messages whose transfer is partly controlled by the receiver. Finally, s**ual selection theory provides a useful theoretical framework for functional examinations of mammalian olfactory communication. Introduction Intraspecifc communication is an essential prerequisite for sociality and allows individuals to make decisions affecting their fitness based on the behavior, morphology, or physiology of conspecifics (Endler 1993; Wilson 1975). Communication, by definition, consists of the transmission of a signal from a sender to a receiver (Endler 1993; Marler 1961; Otte 1974). Many aspects of mammalian sociality, such as spacing, reproductive activity, and competition, are mediated by chemical signals (Albone 1984; Brown and Macdonald 1985). Most previous studies of mammalian olfactory communication focused exclusively on either signaling behavior (e.g., Eisenberg and Kleiman 1972; Johnson 1973; Mykytowycz 1972; Ralls 1971; Thiessen and Rice 1976), the attributes of the signals (e.g., Belcher et al. 1986, 1990; Fuchs et al. 1991; Perret 1992; Schilling and Perret 1987), or the behavioral and physiological effects of the signals on the receiver (see reviews by Albone 1984; Brown and Macdonald 1985; Epple 1986). Comprehensive studies of signal transmission, including emission, reception and the immediate response of the receiver, have rarely been conducted within natural populations (see e.g., Heymann 1998; Hurst 1990a-c; Smale et al. 1990), even though such an approach may disclose additional information about the evolution, design, and constraints of communication in this modality (Alberts 1992; Endler 1993), as demonstrated by insightful an*lyses dealing with other modalities (e.g., Biben et al. 1986; Duncan and Fiske 1979; Marler et al. 1986; Masataka and Biben 1987; Morris and Ryan 1996; Snowdon 1988; Snowdon and Cleveland 1984). Moreover, experimental studies of behavioral and physiological responses to chemical signals have already generated important working hypotheses for such an*lyses (e.g., Clark 1982; Johnston et al. 1997a,b; von Holst and Eichmann 1998; Ziegler et al. 1993). Chemical signals are reliable indicators of individual quality because they are closely linked to physiological conditions (Endler 1993; Guilford and Dawkins 1991; Ha**on 1994). They are therefore probably honest signals that could be used to manipulate the receiver to the benefit of the sender (Dawkins and Krebs 1978; Guilford and Dawkins 1991; Krebs and Dawkins 1984). As with other modalities, the sender of chemical signals controls the information content of the message, and where and under what conditions it will be deposited. Compared to communication in other modalities, however, the exchange of chemical signals is characterized by several idiosyncracies that raise questions about its efficacy and reliability. First, there may be a considerable delay between the emission and perception of a chemical signal, especially for solitary species. Second, many mammalian chemical signals require active investigation by the receiver, which may be beyond the sender's control. Third, counter-marking, which is common and appears to mask the original signal (Johnston et al. 1997a,b), is also beyond the control of the original sender. Thus, a sender may have reduced control over the fate of chemical signals, which may compromise its ability to target a particular conspecific, especially in group-living species, where many potential intended and unintended receivers are continuously present. In addition to methodological difficulties related to this "sender's dilemma,'' at least two other problems have hampered progress towards a better understanding of behavioral and functional aspects of chemical signal transfer among mammals. First, the majority of mammals are solitary and nocturnal, making comprehensive studies of the behavioral aspects of olfactory communication under natural conditions difficult (but see e.g., Hurst 1987, 1989). Second, most studies investigating functional aspects of chemical communication lack a comprehensive theoretical framework. Variation in mammalian scent marking as a function of age, s**, reproductive and social status, location, season, and other factors has been described in many taxa (see e.g., Brown and Macdonald 1985; Epple 1986; Gorman and Trowbridge 1989 for recent reviews), but the suggested ultimate functions of chemical signals were largely based on post hoc correlations. Furthermore, studies focusing on the signals or their effects on the receiver were primarily concerned with signal chemistry and physiological mechanisms and revealed little about their ultimate functions (Albone 1984). It has long been recognized, however, that s**ual selection theory may provide a useful framework for the investigation of mammalian chemical communication (Blaustein 1981; Darwin 1871). This cornerstone of evolutionary theory offers an explanation for the widespread s**ual dimorphism of scent-producing organs, the onset of marking activity with s**ual maturity, the correlation between reproductive hormone levels and scent gland activity, much of the observed variation in scent marking activity, and the physiological effects of many signals. Darwin's (1871) original suggestion for the function of mammalian glands and odors focused on males: "The males are almost always the wooers; and they alone are armed with special weapons for fighting with their rivals. They are provided, either exclusively or in a much higher degree than the females, with odoriferous glands... There is another and more peaceful kind of contest, in which the males endeavour to excite or allure the females by various charms. This may be effected by the powerful odours emitted by the males during the breeding-season; the odoriferous glands having been acquired through s**ual selection.'' It has since been established that females also employ chemical signals to compete with each other and to attract males (Brown 1985; Converse et al. 1995; Drickamer 1992; McClintock 1983). Sexual selection theory should, therefore, provide a useful framework for an an*lysis of the roles of adult males and females as senders and receivers of chemical signals (Blaustein 1981; see also Boake 1991). The exchange of chemical signals plays an important role in the social lives of prosimian primates (Klopfer 1972; Schilling 1979). Ring-tailed lemurs (Lemur catta) are diurnal group-living prosimians from southern Madagascar. Their multi-male, multi-female groups are the largest among prosimians, and their olfactory communication system may be the most intricate among primates (Epple 1985). Males possess highly specialized brachial and antebrachial glands that are used to mark objects in the environment and to impregnate their own tail before waving it at conspecifics. They also deposit scrotal gland secretions, whereas females apply vaginal secretions, and possibly urine, during anogenital marking (Epple 1986; Evans and Goy 1968; Jolly 1966; Mertl-Millhollen 1988). Scent-marking behavior in L. catta is influenced by age, s**, and social status (Kappeler 1990), is frequently observed at the periphery of group home ranges, but is also used to communicate with group members (Mertl-Millhollen 1988; Schilling 1974). All types of chemical signaling are accompanied by conspicuous visual signals, such as bipedal standing and stereotyped movements (Mertl 1976; Ramsay and Giller 1996). Freshly deposited scent marks, which contain information about the s** and individual identity of the sender (Mertl 1975), are often sniffed, and sometimes counter-marked by other group members. For a given scent mark, it is therefore possible to determine the identity of the receiver(s), if any, the nature of the receiver's response to the signal, and the latency period to the response. In this paper, I will quantify different stages of this transmission process and explore non-random patterns in the sequence of subsequent signals in order to provide a first estimate of the efficacy of communication in this modality and to examine the sender's dilemma outlined above. The s** of the sender and receiver will be used as independent variables in this novel methodological approach in order to explore basic predictions of s**ual selection theory about patterns of intra- and inters**ual communication. Methods A group of 26 ring-tailed lemurs at the Duke University Primate Center, Durham, N.C., USA, provided the subjects for this study. They inhabited a 3.5-ha natural-habitat enclosure, where thousands of trees and other parts of the natural vegetation provided unlimited substrates for scent marking, locomotion, and foraging. They were separated from a second L. catta group in an adjacent enclosure by an electrified chain-link fence. All individuals were easily identified by unique neck collars and habituated to the presence of human observers. Further details on the social history and housing conditions of the study group can be found elsewhere (Pereira 1993; Pereira and Kappeler 1997). Scent-marking behavior was observed ad libitum during daily periods of peak activity. An average of 11 scent marks (total n = 214) of each of 9 adult (>2 years) females and 11 adult males served as foci for 10-min observations; the two juvenile males and four juvenile females were not observed. Because of great inter-individual variation in marking frequency (Kappeler 1990), much effort went into obtaining roughly equal numbers of focal scent marks from all individuals. Only scent marks that, to my best knowledge, were deposited on substrates not previously marked by other animals on the same day were used. It must be noted, however, that scent marks may be effective for up to 48 h under laboratory conditions (Mertl 1975). The data were collected between September 1989 and May 1990, including the annual breeding season and the subsequent birth season. I recorded the type of each focal scent mark, as well as the identity of the sender and its nearest neighbor. A stopwatch was started immediately after the deposition of a scent mark to determine the latency to all subsequent interactions with the signal. Whenever a conspecific sniffed or counter-marked a focal scent mark, its identity and the nature of the interaction were recorded. Sniffing, defined as placing the rhinarium within 5 cm of the focal scent mark for at least 1 s, was occasionally a**ociated with licking of the same spot, which may bring the signal in contact with the well-developed vomeronasal organ (Bailey 1978). The responses to the three types of scent marking - male and female anogenital, and male antebrachial marking - were summarized for each type. The proportions of focal scents that were sniffed or counter-marked by at least one conspecific at least once within 10 min were determined and compared by a G-test of independence (Sokal and Rohlf 1981). For counter-marked focal scent marks, I determined the number of first- and second-order transitions by the type of mark or the s** of the animals involved, using all transitions within the 10-min observation periods to investigate patterns of overmarking. The goodness of fit between observed and expected transition frequencies was tested for both zero-and first-order models (Bakeman and Gottman 1986), using a goodness-of-fit G-test. Expected frequencies for zero-order models are calculated by a**uming that events are equiprobable, whereas calculation of expected frequencies for first-order models relies on the observed number of events. The latencies to a response were compared as a function of the type of the focal mark and the s** of the receiver, using two-way ANOVA after log10-transformation. The hypothesis that all group members were equally likely to perceive any scent mark was tested by comparing the observed and expected frequencies with which they interacted with focal scent marks, using a goodness-of-fit G-test. The effects of proximity and visual contact were evaluated by comparing the number of instances in which nearest neighbors were the first receiver of a signal with the number of cases in which they did not interact with a scent, using a binomial test (Siegel and Castellan 1988). The possible effects of social status on signal transmission were examined by comparing the proportion of scents receiving a response between the three highest- and lowest-ranking members of each s** with a G-test. Social status was determined on the basis of decided agonistic interactions (Pereira and Kappeler 1997; Pereira et al. 1990). Male and female ring-tailed lemurs have separate hierarchies and all females dominate all males (Jolly 1966; Kappeler 1993). Frequencies of marking behavior are positively correlated with dominance rank in males, whereas female scent-marking frequencies are independent of social status (Kappeler 1990). Some of these an*lyses were supplemented with focal animal data collected from five adult males and five adult females of the same group over 12 months, including the present study period (see Kappeler 1993). Each focal animal was observed for 15 min twice a week, resulting in 242.5 h of observations. The type of scent mark deposited by a focal animal, its location within the enclosure, and the response of conspecifics to these scent marks during the first minute after their deposition were recorded. This data set was used to examine possible seasonal variation in olfactory communication by comparing marking and response rates across annual seasons, and to describe the spatial distribution of signals. Results Response patterns Of the 214 focal scent marks, 62.1% were received by at least one conspecific within 10 min of deposition. The observed interactions of the receivers with the signals included sniffing, sniffing followed by a counter-mark, or counter-marking without previous sniffing, the latter being observed only once. Of the 123 signals that were sniffed at least once, 109 (88.6%) were also counter-marked at least once. The proportion of received (i.e., sniffed) signals was not significantly different among the three types of focal marks (G = 2.27, 2 df, NS). However, a finer distinction of the possible fates of focal signals revealed that the proportion of counter-marked, sniffed and non-investigated scents differed significantly among the types of focal marks (Fig. 1). All pair-wise post hoc comparisons were significant at P<0.01 (G-test), indicating that the probability of being counter-marked varied among signals. Of those focal marks that were counter-marked at least once within 10 min, female anogenital marks received on average 2.1, male ante-brachial marks 1.7, and male anogenital marks 1.0 counter-mark(s) (Kruskal-Wallis H = 10.35, 2 df, P < 0.01). In only four cases did the sender mark the same spot again within 10 min, and only one animal was observed sniffing its own scent. Temporal aspects The timing of the responses was highly skewed. The frequency distribution of latency periods for first-order transitions revealed that scent marks that were not perceived within about 3 min had a high probability of remaining undetected (Fig. 2). Longer latencies were quite rare, primarily because the group moved on during normal activity. In fact, the median latency between a signal and the subsequent response was only 30 s (interquartile range: 97.5 s). Further an*lysis of the temporal component of the response revealed that the mean latency to the first response was significantly affected by the type of focal mark, but not by the s** of the responding animal (Fig. 3). INSERT IMAGE Fig. 1 Behavioral response to focal scent marks. The proportion of female anogenital (AG) (n = 103), male anogenital (n = 28) and male antebrachial (AG) (n = 83) marks receiving at least one counter-mark (black), sniff (hatched), or no response (open) within the first 10 min of their deposition was significantly different among the types of focal marks INSERT IMAGE Fig. 2 Frequency distribution of latencies for first-order transitions. The number of focal marks sniffed and or counter-marked (n = 136) is shown for each 1-min block during which the first interaction between signal and receiver occurred INSERT IMAGE Fig. 3 Latencies of the response of male and female receivers to three types of scent marks. The mean latencies (±SE) between the deposition of female anogenital (AG), male anogenital, and male antebrachial (AG) marks and the first interaction with the signal by a male (black) or female (hatched) receiver are depicted. The type of focal mark (F2,130 = 4.52, P = 0.013), but not the s** of the receiver (F1,130 = 0.09, NS) had a significant effect on the mean latency. The interaction in this two-way-ANOVA was not significant (F2,130 = 0.69, NS) The sequence of subsequent scent marks was not random. At a descriptive level, I found that the transitional probability for both types of male marks to be followed by another male mark was higher than for all other transitions (Table 1). I found further non-random sequences after comparing the difference between the number of observed and expected first-order transitions. These an*lyses, which revealed significant differences for both zero- and first-order models (Table 2), indicated that males marked more often after females, and females less often after males than expected by chance. The observed frequency of second-order transitions, which describe the sequence of three subsequent signals, was also significantly different from the expected frequency (Table 3), suggesting that the response of an animal was not only influenced by the immediately preceding scent mark and that a scent is not completely masked by a counter-mark. INSERT IMAGE Table 1 Transitional probability matrix. Shown are probabilities for types of marks (lag 0) to be followed by same or other types of marks (lag 1) (f*g female anogenital mark, MAG male anogenital mark, MAB male antebrachial mark) INSERT IMAGE Table 2 First-order transitions. Observed and expected (in parentheses) frequencies of sequences of types of marks (lag 0) to be followed by same or other types of marks (lag 1). Expected values are based on the number of observed marks and equiprobability of transition. The difference between observed and expected values is significant (G = 25.5, 4 df, P < 0.001) (for abbreviations see legend to Table 1) INSERT IMAGE Table 3 Second-order transitions. Observed and expected (in parentheses) frequencies of sequences of types of marks (lag 0 + 1) to be followed by same or other types of marks (lag 2). Types of male marks were combined to increase sample size. Expected values are based on number of observed marks and equiprobability of transition. The difference between observed and expected values is significant (G = 23.6, 3 df, P < 0.01) (M male, F female) Social effects There was no evidence for an audience effect, i.e., scent-marking animals did not appear to mark preferentially in the vicinity of intended receivers. This can be inferred from the observations following a subset (n = 104) of focal scent marks for which relevant data were recorded during the second half of the study. Across all types of these focal marks that received a response (n = 62), the nearest neighbor of the sender was the first to investigate its signal exactly 50% of the time. Investigated female anogenital marks were first inspected by the nearest neighbor in 44% of cases (n = 39, binomial test, NS). For male anogenital and brachial marking, the corresponding percentages were 33 (n = 6, NS) and 71 (n = 17, NS), respectively. Thus, nearest neighbors of senders did not interact with signals more often than other conspecifics. Dominance rank had s**-specific effects on the exchange of olfactory signals. First, the proportion of scent marks deposited by high- and low-ranking males and females, respectively, receiving a response did not differ as a function of rank in either s** (Fig. 4, GFemales = 0.41, 1 df, NS; GMales = 0.33, 1 df, NS). However, rank of female senders and the s** of the receivers were not independent (G = 4.00, 1 df, P < 0.05): males responded more often to scents of high-ranking females and females more often to scents to low-ranking females. There was no such rank effect for male senders, even though scents of low-ranking males were virtually never investigated by females (Fig. 4). A total of 120 responses to focal scents were equally distributed among the 9 females, irrespective of rank (G = 13.4, 8 df, NS; Fig. 5), whereas the top-ranking male responded to other scents much more often than all other males (G = 62.1, 10 df, P < 0.001; Fig. 5). The vast majority of responses of the alpha male consisted of counter-marks. INSERT IMAGE Fig. 4 The effect of sender's rank on the response to their signals. The percentage of scent marks deposited by low- (lo) and high- (hi) ranking males (M) and females (F) that were investigated and/or counter-marked at least once by female (hatched) and male (black) receivers. Only female rank had a significant effect (see text) Focal animal observations Some of these trends obtained from focal scent observations were reflected by the focal animal data. Throughout the year, a total of 395 female scent marks and 647 male scent marks were recorded. The proportion of female marks that was investigated and counter-marked was significantly larger than the proportion of male marks (G = 41.95, 1 df, P < 0.0001; Fig. 6a). The proportions of investigated female marks did not change significantly across months (G = 5.58, 12 df, NS), whereas the proportion of investigated male marks did (G = 26.62, 12 df, P < 0.01). The investigation rates (Fig. 6b), however, changed in parallel with the marking rates (Fig. 7) and peaked during the birth (March) and breeding (October) season for females, whereas there was no obvious correlation with reproductive seasons for males. Finally, the spatial distribution of scent marks was heterogeneous throughout the enclosure, but marking was not concentrated along the boundary with a neighboring group (Fig. 8a). When marking frequencies were corrected for time spent in different quadrants, several areas within the enclosure away from the boundary were marked much more often than expected (Fig. 8b), indicating that there may be hot spots for the exchange of chemical signals within home ranges. Fig. 5 The number and type of responses by females and males towards focal scent marks. A distinction is made between focal scents that were only sniffed/licked (black) and those that were (also) counter-marked (hatched). Both s**es are ordered by rank; females are fully capitalized. 1 represents top rank; all females dominate all males Fig. 6 a The proportion of female (black; total n = 395) and male (open; total N = 647) scent marks deposited by five male and five female focal animals between March 1989 and March 1990 that were investigated by a conspecific within 1 min of their deposition. b Mean number of investigated female (black) and male (open) scent marks per hour during the same time period. Births occurred in March, matings in November Discussion This study provided some new information on the mechanisms and functions of olfactory communication in a group-living mammal. Because it is the first study to quantify these aspects of signal transmission in intact social groups in a naturalistic habitat, the evaluation of some results is necessarily preliminary. Signal transmission The transmission of chemical signals by ring-tailed lemurs was quite efficient: more than 60% were investigated within a few minutes by a group member. Future studies of other taxa will reveal whether this proportion of "wasted'' signals is relatively high or low. These signals are likely wasted because most group members have long pa**ed a scent mark during normal group progression within 10 min and groups rarely return to exactly the same spot on the same or following day. In addition, the median latency of 30 s between the deposition and investigation of a scent also indicates that ring-tailed lemurs monitor the behavior of their group mates closely (see below) and decide soon after the deposition whether or not they will investigate a signal. In wild populations, members of other groups may interact with some of these scents ignored by group mates, however. Fig. 7 The mean number (±SE) of female anogenital, male anogenital and male antebrachial scent marks deposited by five male and five female focal animals between March 1989 and March 1990. Births occurred in March, matings in November Fig. 8 a The spatial distribution of scent marks deposited by five male and five female focal animals between March 1989 and March 1990. x-and y-axis indicate coordinates of 50 x 50 m quadrats within the enclosure (A-H and 2-10, respectively). Marking frequency is shown on the z-axis. The boundary to a neighboring L. catta group ran between quadrats H2 and G9. b Relative marking frequencies after correcting for time spent in each quadrat. Columns with dark tops indicate negative values, i.e., in relation to the time spent, relatively few scent marks were deposited Interestingly, most investigated signals were immediately counter-marked, thereby preventing other potential receivers from receiving the same signal. Experimental studies with golden hamsters (Mesocricetus auratus) and meadow voles (Microtus pennsylvanicus) suggest that males of these species may compete for top position of their scents because females prefer donors of counter-marks in subsequent choice tests over senders of the original signal (Johnston et al. 1994, 1995, 1997a,b). The s** differences in counter-marking demonstrated in this study, and especially the high counter-marking rates of the alpha male, indicate the possibility of a similar function in lemurs. Furthermore, significant second-order transitions observed in this study suggest that at least information about the s** of the two preceding senders is conserved, but experiments are needed to determine whether lemurs can also discriminate between top and bottom scents. The high rate of counter-marking is also relevant to the sender's dilemma. Were these signals intended for the eventual receivers, or were they corrupted by others before the targeted receiver could investigate them? The present observations do not indicate whether and how senders direct their signals to particular group members, because the nearest neighbors of senders were not more likely to investigate freshly deposited scent marks than other animals. The lack of an audience effect in this study is not due to the chosen time window, however, because during regular activity, the group moved on and virtually no animals stayed near a scent mark 10 min after its deposition. It is therefore possible that chemical signals of ring-tailed lemurs are general messages without a particular intended receiver. They may be deposited on a "bulletin board,'' perhaps in front of a cla** of intended receivers, such as adult females or subordinate males, who often rest and travel together (Jolly 1966), and one conspecific completes the act of communication by investigating the signal. Thus, communication in this modality may therefore be ultimately controlled by the receiver. This conclusion is also supported by the various effects of s** of the receivers on qualitative and temporal aspects of the response (see Guilford and Dawkins 1991 for general receiver characteristics). That is to say, characteristics of the receiver explain much of the variation in the fate of signals. Only the unique tail-marking behavior of (male) ringtails may provide a mechanism for direct targeting of chemical signals, because in these agonistic displays, males wave their scent-impregnated tails at a conspecific while standing face to face with them (Jolly 1966). Thus, pending more detailed contexual an*lyses of signaling behavior, it appears that the sender's difficulty in transmitting a signal to a particular group mate is surmounted by conspicuous public signaling (which may contain important information in itself). Olfactory communication differs fundamentally from the acoustic modality, however, because in the latter, many or all group members will receive a given signal. Discrete visual signals, e.g., in the form of facial expressions, which have the potential to be very directed, are virtually lacking from the behavioral repertoire of these lemurs (Pereira and Kappeler 1997), in sharp contrast to many other diurnal primates (e.g., Preuschoft and van Hooff 1995). Instead, ring-tailed lemurs appear to use visual signals in the form of conspicuous body postures and movements to alert their group members to the deposition of a chemical signal (Mertl 1976). Such composite signals, which are common among mammals (Alberts 1992; Johnstone 1996), may, among other things, improve their memorability (Guilford and Dawkins 1991) and detetectability. This particular use of different modalities may also partly reflect recent changes in the activity of these diurnal lemurs, which still exhibit many morphological adaptations to nocturnal life (van Schaik and Kappeler 1996). Signal functions Previous field studies of this and other prosimian primates indicated an important function of chemical signals in between-group communication (Mertl-Millhollen 1986, 1988). One could therefore argue that the observed patterns of signal transmission are an artifact of isolated captive housing. Two observations argue against this possible objection, however. First, wild ringtails also respond in the described manner to scents of group members (Jolly 1966; Mertl 1977; Schilling 1974). Second, members of this captive group marked very often along a fence separating them from a second group in an adjacent enclosure, but there were several other areas within their enclosure where they marked even more often, strongly suggesting a function of these signals in intragroup communication. Nevertheless, it would be interesting to repeat this study in Madagascar to quantify the exchange of chemical signals between groups, as well as to examine potential interindividual differences in sender and receiver roles as a function of where marks are located. As predicted by s**ual selection theory, several aspects of olfactory communication in this species were dependent on the s** of senders and receivers. For example, the distribution of scent-producing organs is s**ually dimorphic, with males having more scent glands than females. Furthermore, scent-marking activity in both s**es begins around puberty, even though some behavioral elements a**ociated with the deposition of scents can already be observed among playing juveniles (Pereira 1993; Pereira and Kappeler 1997). The exchange of chemical signals among males suggested an important function in intras**ual selection. First, scent-marking activity among males is positively correlated with their rank in the linear male hierarchy (Kappeler 1990), and the highest-ranking male is usually also the first one to mate with receptive females and may father isproportionately more offspring (Pereira and Weiss 1991; Sauther 1991). In addition, male marking activity peaked just before the brief annual mating season (see also Evans and Goy 1968; Jolly 1967), possibly reflecting a**ociated endocrinolocical changes (van Horn and Eaton 1979). Second, male scents were most often investigated by other males. Olfactory communication among male house mice is also characterized by high marking and over-marking rates of dominant males and high investigation rates of their scents by subordinates and intruders (Hurst 1990a). In mouse lemurs (Microcebus murinus), odors of dominant males decrease testosterone levels and s**ual activity in other males (Perret 1992). It is therefore conceivable that male scent marks in other lemur species may provide a mechanism for reproductive competition among males as well, especially given the lack of s**ual dimorphism and other adaptations to polygynous mating systems (van Schaik and Kappeler 1996). Third, males investigated female scents much more often than vice versa. With high rates of a**ociated counter-marking, especially by the top-ranking male, males may effectively prevent rivals from sampling this information and communicate their relation to females. Counter-marking may therefore constitute part of interindividual competitive strategies (see Johnston et al. 1994, 1997a). Male squirrel monkeys (Saimiri sciureus) also sniffed marked substrates and genital regions of conspecifics more often than females (Boinski 1993). Among house mice, dominant males counter-marked female scents much more often than any other age-s** cla**, whereas females exhibited comparatively little interest in male scents (Hurst 1990c). Thus, male scents may typically be deposited for the information and/or manipulation of other males, but they may contain important information for female choice in some taxa (see Deutsch and Nefdt 1992). Female scents also got a faster response from both males and females, supporting the notion that female scents are directed towards members of both s**es. In female ring-tailed lemurs, neither age nor social status have an effect on scent-marking behavior (Kappeler 1990). This is a difference from many New World monkeys, where scent-marking behavior of young and subordinate females is frequently suppressed, along with their reproductive activity (Abbott 1989; Goldizen 1987; Heymann 1998). Females also investigated and counter-marked scents deposited by other females, especially those of low rank, more often than male scents. Among female house mice, breeding individuals also showed the strongest counter-marking responses to other females' urine (Hurst 1990b). In ringtails, this pattern may be related to the intense competition among females, in particular the phenomenon of targeted aggression, where females, often low ranking, target other females for persistent aggression, frequently resulting in serious injuries and/or eviction of the target (Pereira 1993; Vick and Pereira 1989). Females may monitor physiological changes that could predict targeted aggression and/or mask the scents of low rankers by counter-marking. Unfortunately, the present data were insufficiently detailed to address these questions, but these testable predictions could provide an interesting focus for future studies. Finally, females may signal changes in reproductive condition to males. This likely function of female scents was suggested by the annual variation in marking activity, the male response to female scents, as well as other aspects of male behavior, such as genital licking of females, which also peaked in frequency just before and during the mating season (Evans and Goy 1968). Such a function has also been demonstrated or suggested in other primates (Converse et al. 1995; Fornasieri and Roeder 1992; French et al. 1989; Heistermann et al. 1989; Kappeler 1988; Ziegler et al. 1993) and appears to be a basic mammalian trait (Eisenberg and Kleiman 1972; Ralls 1971). Thus, the behavior of both male and female ring-tailed lemurs in their roles as senders and receivers of olfactory signals suggests that primary functions of female olfactory signals in intragroup communication may be in male attraction and competition with other females, whereas male pheromones may function primarily in intras**ual competition. It must be emphasized, however, that this first an*lysis of general patterns did not demonstrate any proximate or ultimate functions of olfactory signals. Both more detailed etho-endocrinological studies and experiments in this species and comparative data from other taxa are now needed to refine and test these working hypotheses. Acknowledgements Thanks to the staff and directors of the DUPC for providing a unique environment for lemur research. I also want to thank Anne Nacey for her help with data collection and for inspiring discussions, Michael Pereira for companionship in the field and for teaching me how to observe ringtails, and Peter Klopfer for his guidance and advice. Fritz Trillmich, Eckhard Heymann and three anonymous reviewers improved an earlier version of this man*script with their suggestions. This is Duke University Primate Center publication no. 664. References Abbott DH (1989) Social suppression of reproduction in primates. In: Standen V, Foley RA (eds) Comparative socioecology. Blackwell, Oxford, pp 285-304 Alberts AC (1992) Constraints on the design of chemical commu nication in terrestrial vertebrates. Am Nat 139:S62-S89 Albone ES (1984) Mammalian semiochemistry: the investigation of chemical signals between mammals. Wiley, New York Bailey K (1978) Flehmen in the ring-tailed lemur (Lemur catta). Behaviour 65:309-319 Bakeman R, Gottman JM (1986) Observing interaction: an introduction to sequential an*lysis. Cambridge University Press, Cambridge, UK Belcher AM, Smith AB, Jurs PC, Lavine B, Epple G (1986) an*lysis of chemical signals in a primate species (Saguinus fuscicollis): use of behavioral, chemical, and pattern recognition methods. J Chem Ecol 12:513-531 Belcher AM, Epple G, Greenfield KL, Richards LE, Smith AB (1990) Proteins: biologically relevant components of the scent marks of a primate (Saguinus fuscicollis). Chem Sens 15:431- 446 Biben M, Symmes D, Masataka N (1986) Temporal and structural an*lysis of aliative vocal exchanges in squirrel monkeys (Saimiri sciureus). Behaviour 29:259-273 Blaustein AR (1981) Sexual selection and mammalian olfaction. Am Nat 117:1006-1010 Boake C (1991) Coevolution of senders and receivers of s**ual signals: genetic coupling and genetic correlations. Trends Ecol Evol 6:225-227 Boinski S (1993) Olfactory communication among Costa Rican squirrel monkeys: a field study. Folia Primatol 59:127-136 Brown JE (1985) The rodents I: effects of odours on reproductive physiology (primer effects). In: Brown RE, MacdonaldDW(eds) Social odours in mammals. Clarendon, Oxford, pp 245-344 Brown RE, Macdonald DW (1985) Social odours in mammals. Clarendon, Oxford Clark AB (1982) Scent marks as social signals in Galago cra**icaudatus. I. Sex and reproductive status as factors in signals and responses. J Chem Ecol 8:1133-1151 Converse L, Carlson A, Ziegler T, Snowdon C (1995) Communication of ovulatory state to mates by female pygmy marmosets, Cebuella pygmaea. Anim Behav 49:615-621 Darwin C (1871) The descent of man and selection in relation to s**. Murray, London Dawkins R, Krebs JR (1978) Animal signals: information or manipulation? In: Krebs JR, Davies NB (eds) Behavioural ecology: an evolutionary approach. Blackwell, Oxford, pp 282-309 Deutsch JC, Nefdt RJC (1992) Olfactory cues influence female choice in two lek-breeding antelopes. Nature 356:596-598 Drickamer L (1992) Oestrous female house mice discriminate dominant from subordinate males and sons of dominant from sons of subordinate males by odour cues. Anim Behav 43:868-870 Duncan S, Fiske DW (1979) Dynamic patterning in conversation. Am Sci 67:90-98 Eisenberg JF, Kleiman DG (1972) Olfactory communication in mammals. Annu Rev Ecol Syst 3:1-32 Endler J (1993) Some general comments on the evolution and design of animal communication systems. Phil Trans Roy Soc Lond B 340:215-225 Epple G (1985) The primates I: order Anthropoidea. In: Brown J, Macdonald, D (eds) Social odours in mammals. Clarendon, Oxford, pp 739-769 Epple G (1986) Communication by chemical signals. In: Mitchell G, Erwin J (eds) Comparative primate biology 2A. Liss, New York, pp 531-580 Evans CS, Goy RW (1968) Social behaviour and reproductive cycles in captive ringtailed lemurs (Lemur catta). J Zool (Lond) 156:181-197 Fornasieri I, Roeder JJ (1992) Marking behaviour in two lemur species (L. fulvus and L. macaco): relation to social status, reproduction, aggression and social change. Folia Primatol 59:137-148 French JA, Abbott DH, Snowdon CT (1989) The effect of social environment on estrogen secretion, scent marking, and socio-s**ual behavior in tamarins (Saguinus oedipus). Am J Primatol 6:155-167 Fuchs E, Rosenbusch J, Anzenberger G (1991) Urinary protein patterns reflect social rank in male common marmosets (Callithrix jacchus). Folia Primatol 57:177-180 Goldizen AW (1987) Tamarins and marmosets: communal care of offspring. In: Smuts BB, Cheney DL, Seyfarth RM, Wrangham RW, Struhsaker TT (eds) Primate societies. University of Chicago Press, Chicago, pp 34-43 Gorman ML, Trowbridge BJ (1989) The role of odor in the social lives of carnivores. In: Gittleman JL (eds) Carnivore behavior, ecology, and evolution. Cornell University Press, Ithaca, NY, pp 57-88 Guilford T, Dawkins MS (1991) Receiver psychology and the evolution of animal signals. Anim Behav 42:1-14 Ha**on O (1994) Cheating signals. J Theor Biol 167:223-238 Heistermann M, Kleis E, Pröve E, Wolters HJ (1989) Fertility status, dominance and scent-marking behavior of family-housed cotton-top tamarins (Saguinus oedipus) in the absence of their mothers. Am J Primatol 18:177-189 Heymann EW (1998) Sex differences in olfactory communication in a primate, the moustached tamarin, Saguinus mystax (Callitri-chinae). Behav Ecol Sociobiol 43 (in press) Holst D von, Eichmann F (1998) Sex-specific regulation of marking behavior by s** hormones and conspecific scent in tree shrews (Tupaia belangeri). Physiol Behav 63:157-164 Horn RN van, Eaton GG (1979) Reproductive physiology and behavior in prosimians. In: Doyle GA, Martin RD (eds) The study of prosimian behavior. Academic Press, New York, pp 79-122 Hurst JL (1987) The functions of urine marking in a free-living population of house mice, Mus domesticus Rutty. Anim Behav 35:1433-1442 Hurst JL (1989) The complex network of olfactory communication in populations of wild house mice, Mus domesticus Rutty: urine 420 marking and investigation within family groups. Anim Behav 37:705-725 Hurst JL (1990a) Urine marking in populations of wild house mice Mus domesticus Rutty. I. Communication between males. Anim Behav 40:209-222 Hurst JL (1990b) Urine marking in populations of wild house mice Mus domesticus Rutty. II. Communication between females. Anim Behav 40:223-232 Hurst JL (1990c) Urine marking in populations of wild house mice Mus domesticus Rutty. III. Communication between the s**es. Anim Behav 40:233-243 Johnson RP (1973) Scent marking in mammals. Anim Behav 21:521-535 Johnston RE, Chiang G, Tung C (1994) The information in scent over-marks of golden hamsters. Anim Behav 48:323-330 Johnston RE, Munver R, Tung C (1995) Scent counter marks: selective memory for the top scent by golden hamsters. Anim Behav 49:1435-1442 Johnston RE, Sorokin ES, Ferkin MH (1997a) Female voles discriminate males' over-marks and prefer top-scent males. Anim Behav 54:679-690 Johnston RE, Sorokin ES, Ferkin MH (1997b) Scent counter-marking by male meadow voles: females prefer the top-scent male. Ethology 103:443-453 Johnstone RA (1996) Multiple displays in animal communication: `backup signals' and `multiple messages'. Phil Trans R Soc Lond B 351:329-338 Jolly A (1966) Lemur behavior. University of Chicago Press, Chicago Jolly A (1967) Breeding synchrony in wild Lemur catta. In: Altman SA (ed) Social communication among primates. University of Chicago Press, Chicago, pp 3-14 Kappeler PM (1988) A preliminary study of olfactory behavior of captive Lemur coronatus during the breeding season. Int J Primatol 9:135-146 Kappeler PM (1990) Social status and scent marking behaviour in Lemur catta. Anim Behav 40:774-776 Kappeler PM (1993) Variation in social structure: the effects of s** and kinship on social interactions in three lemur species. Ethology 93:125-145 Klopfer PH (1972) Communication in prosimians. In: Sebeok TH (eds) How animals communicate. Indiana University Press, Bloomington, pp 841-850 Krebs JR, Dawkins R (1984) Animal signals: mind reading and manipulation. In: Krebs JR, Davies NB (eds) Behavioural ecology: an evolutionary approach. Blackwell, Oxford, pp 380-402 Marler P (1961) The logical an*lysis of animal communication. J Theor Biol 1:295-317 Marler P, Dufty A, Pickert R (1986) Vocal communication in the domestic chicken. II. Is a sender sensitive to the presence and nature of a receiver? Anim Behav 34:194-198 Masataka N, Biben M (1987) Temporal rules regulating affiliative vocal exchanges of squirrel monkeys. Behaviour 101:311-319 McClintock MK (1983) Pheromonal regulation of the ovarian cycle: enhancement, suppression and synchrony. In: Vandenberg JG (ed) Pheromones and reproduction in mammals. Academic Press, New York, pp 113-150 Mertl AS (1975) Discrimination of individuals by scent in a primate. Behav Biol 14:505-509 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 (1986) Territorial scent marking by two sympatric lemur species. In: Duvall M, Mueller-Schwarze D, Silverstein S (eds) Chemical signals in vertebrates, vol 4. Ecology, evolution, and comparative biology. Plenum, New York, pp 647-652 Mertl-Millhollen AS (1988) Olfactory demarcation of territorial but not home range boundaries by Lemur catta. Folia Primatol 50:175-187 Morris M, Ryan M (1996) Sexual difference in signal-receiver coevolution. Anim Behav 52:1017-1024 Mykytowycz R (1972) The behavioural role of the mammalian skin glands. Naturwissenschaften 59:133-139 Otte D (1974) Effects and functions in the evolution of signaling systems. Annu Rev Ecol Syst 5:385-417 Pereira ME (1993) Agonistic interaction, dominance relation, and ontogenetic trajectories in ringtailed lemurs. In: Pereira ME, Fairbanks LA (eds) Juvenile primates: life history, development, and behavior. Oxford University Press, New York, pp 285-305 Pereira ME, Kappeler PM (1997) Divergent systems of agonistic relationship in lemurid primates. Behaviour 134:225-274 Pereira ME, Weiss ML (1991) Female mate choice, male migration, and the threat of infanticide in ringtailed lemurs. Behav Ecol Sociobiol 28:141-152 Pereira ME, Kaufman R, Kappeler PM, Overdorff DJ (1990) Female dominance does not characterize all of the Lemuridae. Folia Primatol 55:96-103 Perret M (1992) Environmental and social determinants of s**ual function in the male lesser mouse lemur (Microcebus murinus). Folia Primatol 59:1-25 Preuschoft S, Hooff J van (1995) h*mologizing primate facial displays: a critical review of methods. Folia Primatol 65:121-137 Ralls K (1971) Mammalian scent marking. Science 171:443-449 Ramsay N, Giller P (1996) Scent-marking in ring-tailed lemurs: responses to the introduction of "foreign" scent in the home range. Primates 37:13-23 Sauther ML (1991) Reproductive behavior of free-ranging Lemur catta at Beza Mahafaly Special Reserve, Madagascar. Am J Phys Anthropol 84:463-477 Schaik CP van, Kappeler PM (1996) The social systems of gregarious lemurs: lack of convergence with anthropoids due to evolutionary disequilibrium? Ethology 102:915-941 Schilling A (1974) A study of marking behavior in Lemur catta. In: Martin RD, Doyle GA, Walker AC (eds) Prosimian biology. Duckworth, London, pp 347-362 Schilling A (1979) Olfactory communication in prosimians. In: Doyle GA, Martin RD (eds) The study of prosimian behavior. Academic Press, New York, pp 461-542 Schilling A, Perret M (1987) Chemical signals and reproductive capacity in a male prosimian primate (Microcebus murinus). Chem Sens 12:143-158 Siegel S, Castellan NJ (1988) Nonparametric statistics for the behavioral sciences. McGraw-Hill, New York Smale L, Pedersen JM, Block ML, Zucker I (1990) Investigation of conspecific male odours by female prairie voles. Anim Behav 39:768-774 Snowdon CT (1988) Communication as social interaction: its importance in ontogeny and adult behavior. In: Todt D, Goedeking P, Symmes D (eds) Primate vocal communication. Springer, Berlin Heidelberg New York, pp 108-132 Snowdon CT, Cleveland J (1984) "Conversations" among pygmy marmosets. Am J Primatol 7:15-20 Sokal RR, Rohlf FJ (1981) Biometry. Freeman, New York Thiessen D, Rice M (1976) Mammalian scent gland marking and social behavior. Psychol Bull 83:505-539 Vick LG, Pereira ME (1989) Episodic targeting aggression and the histories of Lemur social groups. Behav Ecol Sociobiol 25:3-12 Wilson EO (1975) Sociobiology. Belknap, Cambridge, Ma** Ziegler TE, Epple G, Snowdown CT, Porter TA, Belcher AM, Küderling I (1993) Detection of the chemical signals of ovulation in the cotton-top tamarin, Saguinus oedipus. Anim Behav 45:313-322 Communicated by F. Trillmilch