Behavioral Ecology Vol. 12 No. 6: 698-705
© 2001 International Society for Behavioral Ecology
Scent-marking by male mice under the risk of predation
a Institute of Zoology, Zoological Society of London, Regent's Park, London NW1 4RY, UK b Department of Psychology, University College London, Gower Street, London WC1E 6BT, UK c Evolution and Behavior Research Group, Dept of Psychology, University of Newcastle, Newcastle-upon-Tyne NE1 7RU, UK
Address correspondence to S.C. Roberts, who is now at the Evolution and Behavior Research Group, Department of Psychology, University of Newcastle, Newcastle-upon-Tyne NE1 7RU. E-mail: craig.roberts{at}ncl.ac.uk .
Received 26 June 2000; revised 28 November 2000; accepted 5 February 2001.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The use by predators of scent marks made by potential prey is a largely unexplored potential cost of olfactory signaling. Here we investigate how animals that differ in their investment in scent-marking respond to simulated predation risk, by comparing the willingness to approach and counter-mark the scent marks of a competitor in the presence or absence of predator odor. We aimed to test whether animals that invest heavily in scent-marking, and which may thus be more vulnerable to eavesdropping predators, will either (1) take greater risks to counter-mark the competitor's scent or (2) reduce or abandon scent-marking. Using outbred male laboratory mice, Mus musculus, we show that, in the absence of predators, individuals which initially scent-mark at high frequency (high markers) approach the competitor's scent marks more quickly and spend more time in counter-marking than those which initially invest relatively little (low markers). In a sib-sib experimental design, simulated presence of predation risk (urine of ferrets, Mustela putorius furo) caused both kinds of individual to approach the competitor's marks more slowly, but high markers approached more quickly than low markers and spent more time in the vicinity of the competitor's marks. Only high markers significantly reduced their overmarking of the competitor's scent. These results suggest (1) that there is a unique danger inherent to scent-marking at high frequencies and (2) that high-marking males were prepared to accept increased costs of intrasexual competition in order to reduce the risk of predation. Further tests using the scent of naked mole-rats, Heterocephalus glaber, showed that these effects were not elicited simply by an unfamiliar odor. We discuss reasons for the observed difference in response to predation risk between the groups, and the implications of these results for counter-selection on scent-marking strategies.
Key words: mouse, olfaction, signal cost, status signals.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Status signals are expected to be costly to signalers if they are reliable, cheat-proof indicators of quality (Grafen, 1990
; Zahavi,
1975). The currencies in which costs are expressed vary widely
across signal modalities and taxa. An important potential cost of signal
emission is that signalers may inadvertently expose their location to
eavesdropping parasites and predators
(McGregor, 1993;
Stoddard, 1999). For example,
dipteran parasitoids use cricket song to locate potential larval hosts
(Cade, 1975) and vertebrate
predators use frog (Tuttle and Ryan,
1981) and insect vocalizations
(Bailey and Haythornthwaite,
1998; Belwood and Morris,
1997; Sakaluk and Belwood,
1984) to localize displaying males. Counter-selection on signalers
is thought to induce individual variation in response to predation risk
(Endler, 1992). Males may
follow alternative signaling strategies in this context, such as adopting
silent satellite roles (Cade,
1975) or varying signal form
(Bailey and Haythornthwaite,
1998; Stoddard,
1999; Zuk et al.,
1993). These strategies may help to reduce vulnerability to
predation but may also reduce the benefits associated with signaling,
particularly success in intrasexual competition or attractiveness to mates
(Belwood and Morris, 1997;
Burk, 1982;
Endler, 1992).
While the use of olfactory social signals to detect prey is less well
documented than that of acoustic signals, recent work has shown that some
raptors, such as the kestrel (Falco tinnunculus,
Viitala et al., 1995) and
rough-legged buzzard (Buteo lagopus,
Koivula and Viitala, 1999),
use the scent marks of voles as a cue to likely prey density and that they
prefer to hunt in areas with high densities of scent marks. However, we are
not aware of any study which has examined individual variations in
scent-marking behavior in relation to predation risk. As scent-marking, like
acoustic signaling, typically functions both in intrasexual competition
between males (Brown and Macdonald,
1985; Gosling,
1982,
1990;
Gosling and Roberts, 2001) and
in advertising male quality to potential mates
(Rich and Hurst, 1999),
changes in signaling behavior in response to a higher risk of predation should
vary between males of different status.
High-quality males usually scent-mark at high frequency, constructing a
dense and often complex system of marks within their territory or area of
dominance (Gosling, 1981;
Roberts and Lowen, 1997).
Since these marks may reveal (indeed signal) predictable patterns of behavior
and movement, we suggest that high quality territorial or dominant males will
be more vulnerable in the presence of predators than males investing
relatively little in scent-marking, such as sympatric nonterritorial or
subordinate males. We therefore aimed to test two alternative hypotheses that
suggest how males that scent-mark at high frequency might respond to the
presence of a predator. First, males that adopt a high-marking strategy may be
prepared to take relatively greater risks in signaling than low-marking males
(cf. Trivers, 1972;
Williams, 1966). Second, they
might reduce or abandon scent-marking if the benefits of signaling are
outweighed by high predation risk (Belwood
and Morris, 1997; Zuk et al.,
1993).
As there is typically a delay between signal emission and reception in
olfactory signaling, the main mechanism through which receivers can positively
identify the signaler is thought to be scent-matching, in which competitors or
mates match the odor from scent marks with the odor of conspecifics they
encounter (Gosling, 1982;
Gosling and McKay, 1990). It
is thus crucial that signalers maintain their scent marks in such a way that
maximizes the success of matching
(Gosling, 1986;
Gosling and Roberts, 2001;
Roberts and Lowen, 1997). This
is achieved both by replenishing their own scent marks on a regular basis and
by counter-marking any scent deposited by competitors within their territory
or area of dominance (Rich and Hurst,
1999; Roberts,
1998). In small rodents, males of high status are known to both
scent-mark (Desjardins et al.,
1973; Hurst, 1990)
and counter-mark to a greater extent than individuals of lower status
(Hurst, 1990;
Johnson, 1973). These trends
increase success in maintaining status
(Gosling and McKay, 1990;
Gosling et al., 2000), deter
challenges from poor quality individuals (Gosling et al.,
1996a,
1996b;
Hurst et al., 1994) and
enhance the signaler's attractiveness to mates
(Johnston et al., 1997b;
Rich and Hurst, 1999).
In this article, we examine how animals of different quality, varying only in the amount they invest in scent-marking, respond to the risk of predation. We present results from an experiment in which we simulate territorial intrusions by transferring between cages the scent marks of male laboratory mice, Mus musculus, and test the willingness of males to counter-mark the intruder's scent, either in the presence or absence of the scent of a predator (ferret, Mustela putorius furo). We interpret the results in the light of the two proposed hypotheses for the response of individuals at relatively higher risk and discuss the implications of these findings for ideas about the evolution of olfactory signals.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Husbandry and techniques
We used an outbred strain of laboratory mice (TO, purchased from B&K Universal Ltd., Hull, UK), in which there is large variation in scent-marking rates between individuals and where this variation arises early in ontogeny (Collins et al., 1996; Gosling et al., 2000
). While the use of laboratory mice in an experiment to
investigate predation risk might be criticized on the grounds that subjects
would not have prior experience of predators, preliminary observations
suggested that, at least in this outbred strain, mice responded adaptively:
when two tubes (one clean, one containing predator odor) were presented
simultaneously, mice avoided the one containing predator scent but freely
entered the clean tube. We established breeding pairs which were housed in MB1
cages (45 x 28 x 13 cm) and fed ad libitum breeder diet
RM3 and water, under a reversed 12:12 light:dark cycle at constant temperature
(21 ± 2°C). During the dark phase of the cycle, red lights were
used to allow observation without disrupting mouse activity cycles. Sires were
removed two weeks after pairing. Twenty litters were obtained, born within a
5-day span. At age 3 days, each litter was culled to a 3:2 M:F sex ratio to
control for potential influences of social environment on subsequent
competitive strategies (Collins et al.,
1997; Mendl and Paul,
1990). Litters were weaned at 3 weeks of age and female sibs were
removed. After weaning, males were housed in sibling groups (MB1 cages,
maintenance diet RM1) for a further 2 weeks, during which dominance
relationships were assessed. We first used two perforated translucent plastic
(perspex) dividers to separate the three sibs within their cage for 30 min.
Litters were then videotaped from above for 15 min after removal of the
dividers, which stimulated agonistic interactions and facilitated the
assessment of relationships (Brain and
Kamal, 1989). We recorded fighting behavior, biting, persistent
following of one individual by another, and submissive or defensive posturing.
Dominance status was attributed on the basis of the frequency with which
individuals displayed these behaviors to their siblings and the outcome of
such interactions.
At 5 weeks of age, the three male sibs were isolated and, during weeks 6
and 8, scent-marking rates of all males were measured. A plastic block (5
x 5 x 4 cm) covered by clean Benchkote absorbent paper was placed
in the cage for 1 h. Mice regularly scent-mark novel objects in their
environment, enabling us to assess individual variation in the intensity of
scent-marking. The 1-h test period has been found to be an appropriate measure
of scent-marking variation between individuals while minimizing the degree of
overlap between successive deposits
(Collins et al., 1997;
Gosling et al., 2000). We used
ninhydrin to stain the scent marks deposited by males on the paper, and
quantified the percent cover of the marks using an overlaid transparent grid.
These data were arcsine-transformed before analysis
(Sokal and Rohlf, 1995). Mean
litter marking rates were calculated and litters were divided into two groups
based on these means (10 litters which marked at high rates and 10 which
marked at low rates). The marking rate of the high-marking group (mean 34%,
range 23-43) was almost three times higher than that of the low-markers (mean
12%, range 4-21; t18 = 8.21, p <.001). We used
litter means (the mean of the three sibs' marking rates) because there is
typically significantly more variation between than within litters
(Collins et al., 1997,
unpublished data), and because of our sib-sib experimental design. We divided
mice into treatment groups on the basis of their scent-marking rates within
litters, rather than dominance status, because our interest was in the
response to predation risk specifically in the context of levels of signal
investment. In addition, we controlled for dominance status within litters by
allocating equal numbers of dominants and subordinates to each group (sibs of
intermediate status from each litter were set aside for control trials using
mole-rat scent, see below) and by housing males in isolation for 6 weeks
before carrying out this experiment. Isolated male mice are known to behave
similarly to dominant males (Bronson and
Marsden, 1973; Corridi et al.,
1993; Parmigiani et al.,
1981). We later checked whether these measures had worked, by
testing for residual effects of within-litter status on responses to intruder
and predator scent, but we found no significant effects (p >.1 in
all comparisons).
Experimental design
Each mouse was transferred to its experimental cage 1 week before
observations began, when the mice were 12 weeks old. These were large plastic
cages (RC1, 56 x 38 x 18 cm) with a plywood inner framework of
walls that divided the cage into two sections
(Figure 1). One section
contained a wooden nestbox and allowed access to food and water, while the
other comprised an open area in which two scent-marking blocks were fixed in
position (termed the experimental section). The dividing wall between the two
sections was pierced by two small holes to allow olfactory contact between
sections. Direct access between sections was possible only through a 15 cm
transparent plastic tunnel (5 cm in diameter); a plywood wall running
alongside the tunnel prevented mice from viewing the experimental section
until they emerged from it.
|
In each of the trials, mice were placed in the nestbox while its two
scent-marking blocks were removed and replaced by one clean block and one
previously marked by an unfamiliar, "donor" male (additional
trials were run on the day before observation, without any introduced scent,
in order to familiarize males with replacement of the blocks). Donor males
were 2 weeks older than the subjects and were bought directly from the
supplier at 5 weeks old and housed singly (M2 cages, 20 x 10 x 13
cm). Of 85 potential donors whose scent-marking rates were measured, the half
(42 males) which marked at the highest rates were selected for use in the
trials, to ensure that simulated intruders were of high quality and that the
introduced blocks were comprehensively scent-marked. Scent marks were
collected from donors by introducing scent-marking blocks into their cages for
1 h. They were collected 24 h before use, frozen overnight at -20°C in
sealed plastic bags and placed at room temperature for 30 min before use. This
procedure minimized the loss of volatile components of the scent marks and its
possible effects on subject behavior
(Roberts, 1998). Blocks were
observed under ultra-violet light before use, to confirm that they had been
scent-marked by the donor. The respective positions in the cage of the clean
and competitor-marked blocks were alternated between trials.
In addition to the competitor's scent, half the mice in each of the high- and low-marking groups (one sib from each sib pair) were exposed to the odor of a potential predator. Fresh sawdust (replaced daily) from the floor of an open-plan enclosure containing five ferrets was collected on the day of the trials. A standard volume (100 ml) was placed inside the tunnel and sprinkled around the substrate of the experimental section at the same time as the blocks were replaced. To ensure that this was a novel odor, trials in which ferret scent was used were carried out in a different room from those with only competitor scent, and mice were brought into the room immediately before the trial began. Using this sib-sib design, 10 males received only competitor scent and 10 received both competitor and predator scents, in each of the high and low-marking groups.
Trials began when the nest-box door was removed. Individuals were then observed for the following 30 min. During this time, the latency to passing through the tunnel and time spent in the experimental section were recorded (one low-marker did not pass through the tunnel when exposed to both competitor and predator odor). In addition, the use of the scent-marking blocks was noted, including the latency to visit each of the blocks, and the scent-marking rate and time spent on them (the last two measures were divided by the time spent in the experimental section, to control for time spent elsewhere). Comparisons between the time spent on each of these blocks indicated the degree of interest in the competitor's scent.
Scent-marking rates were measured in two ways. First, direct measurement
was taken by scoring the percent coverage on the clean block, using the same
methods as previously described. This measured the amount that males marked in
the immediate vicinity of the introduced marks (counter-marking,
Johnston et al., 1997b;
Rich and Hurst, 1999). Second,
time spent on the competitor-marked block was used to estimate the amount of
over-marking of the competitor's scent by the subject. This could not be
determined directly, as it is known from other studies that rodents often
place scent marks on top of, or overlapping, competitor's marks
(Ferkin, 1999;
Johnston et al., 1997a). We
verified that this was a reasonable surrogate measure of direct overmarking
rate by the following analyses. First, we compared the amount of time spent on
clean blocks with the resulting direct measure of marking rate on the same
block during the familiarization trial, in which neither competitor nor
predator odor was present. Time spent on the block was found to be positively
related to the measurement of scent-marking rate (Pearson r =.377,
n = 40, p =.018). Second, and more importantly, the same
relationship was found to be much stronger during experimental conditions,
when odors of both competitor and predator were present (r =.643,
n = 19, p <.01). Last, we also found a positive
relationship between the time spent on the competitor-marked block and the
scent-marking rate as measured on the clean block
(Figure 2; r =.782,
n = 19, p <.001). Here, we present both measures of the
amount of counter-marking in our analyses in order to compare the response to
the competitor's scent between the two groups of mice in the presence or
absence of predator scent.
|
To ensure that the recorded responses were not simply due to the introduction of an unfamiliar scent, we conducted further trials in which the third male sibling (of intermediate status, not used above) was exposed to the scent of the naked mole-rat, Heterocephalus glaber, rather than the predator. In addition to the competitor's scent, the mole-rat scent was introduced in the form of soiled sawdust and scattered in the tunnel and floor of the cage in the same way as the predator scent. The design was identical in all other respects to the trials described above.
Analysis
Data were recorded using Psion Organisers (LZ64) and Observer for Windows
v.3, and analyzed using SPSS v.7 for Windows. Data were checked for normality
before analysis. Measures of latency were not normally distributed and we thus
used nonparametric tests to compare latencies between sibs (Wilcoxon signed
ranks tests) and between high- and low-marking groups (Mann-Whitney U
tests). In other comparisons, we used either paired or independent t
tests. We used Levene's test for equality of variance before carrying out
t tests and where these are unequal the degrees of freedom were
adjusted accordingly; outliers (more than two standard deviations from the
mean) were discounted. Scent-marking rates (which were skewed towards zero)
and other proportional data were arcsine-transformed before analysis. All
tests are two-tailed.
Ethical note
Following Huntingford
(1984), predation risk was
simulated using predator scent in preference to using live predators. Pilot
experiments were carried out before data recording began to ensure that mice
were able to move freely throughout the cage and use the feeding and drinking
stations in the presence of the predator scent. During determination of
dominance relationships, litters were observed during videotaping to check for
injuries and mice would have been separated if injuries occurred
(within-litter aggression typically occurs at low levels compared to adult
mice). However, no injuries were seen. Cardboard tunnels were provided to
offer routes of escape from dominant sibs and for environmental
enrichment.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Effects of predator odor on mouse movement
Presence of predator odor influenced the response of male mice to the competitor's odor in a number of ways. In both high- (Wilcoxon signed ranks test, T = 0, n = 10, p <.01) and low-marking subjects (T = 0, n = 10, p <.01), males which encountered predator odor took longer to pass through the tunnel than their sibs which did not encounter predator odor (Figure 3a). In addition, low-marking mice spent less time than their sibs beyond the tunnel (on the experimental side of the arena) when they encountered predator scent (Figure 3b; paired t test, t9 = 5.04, p =.001). However, there was no comparable difference among high-marking males (t7 = 0.612, ns). Comparing between groups, high-markers passed more quickly through the tunnel than low-markers, both when predator scent was present (Mann-Whitney test, U = 18, n1 = n2 = 10, p <.02) and absent (U = 14, n1 = n2 = 10, p <.01). High- and low-marking males spent similar amounts of time (around 80%) on the experimental side when predator odor was absent (independent samples t test, t18 = 1.96, ns). When it was present, however, high-markers spent more time there than low-markers (t18 = 3.19, p <.01).
|
Overmarking the competitor's scent
Both high- (T = 2, n = 9, p <.02) and
low-marking (T = 0, n = 10, p <.01) males which
encountered predator scent took longer to visit the competitor-marked block
for the first time than their sibs which did not encounter predator scent
(Figure 4a). Comparing between
groups, high-markers were always quicker than low-markers to visit the marked
block, whether predator scent was absent (U = 8, n1 =
n2 = 10, p =.001) or present (U = 17,
n1 = 9, n2 = 10, p <.05).
|
The presence of predator scent did not influence the amount of time that low-markers spent on the competitor-marked block (Figure 4b; t8 = 0.002, ns). In contrast, high-markers that encountered predator scent spent significantly less time on the competitor-marked block than their sibs that did not encounter predator scent (t9 = 5.03, p =.001). These results indicate that the presence of predator scent caused high-markers, but not low-markers, to reduce their investment in overmarking. As a result of this difference in response, high-markers spent significantly less time on the competitor-marked block than low-markers in the presence of predation risk (t11.54 = 2.77, p <.02), even though there was no difference between the two groups when predator scent was absent (t18 = 1.40, ns).
Scent-marking on the clean block
Patterns in the amount of time spent on the clean block were different from
those on the marked block, but, since mice spent (on average) six times longer
on the marked block (Figure 5),
clean block marking rates are less representative of the response to the two
stimuli presented. Mice spent less time on the clean block both when predator
odor was present (Figure 5;
high-markers, t9 = 5.02, p =.001; low-markers,
t8 = 3.93, p <.01) and absent (high-markers,
t9 = 16.91, p <.0001; low-markers,
t9 = 4.40, p <.01). Despite this, the clean
block enabled direct measurement of scent-marking rates during the experiment.
Low-marking mice that encountered predator scent marked at higher rates on the
clean block than their sibs that did not encounter predator scent
(Table 1). In contrast, the
presence of predator scent did not change marking rates of the high-marking
sibs on the clean block. There was no difference between the high- and
low-marking groups in the total amount of marking recorded on the clean block
during the experiment (low-markers, 10.7 ± 2.5%, high-markers, 13.3
± 2.3%; t18 = 1.06, ns), but differences became
apparent when we controlled for the amount of time spent near the blocks, on
the experimental side of the arena. As expected, high-markers marked at higher
rates on the clean block than low-markers, at least in the absence of predator
scent (t17 = 2.17, p <.05). However, the fact
that the two groups responded differently to predator scent produced the
somewhat surprising result that, under the unusual conditions of predator
scent being present, low-markers in fact marked at higher rates on the clean
block than high-markers (t17 = 2.27, p
<.05).
|
|
Effects of unfamiliar, nonpredator scent
The response of mice to the presence of unfamiliar mole-rat scent, compared
with the presence or absence of predator scent in the above trials, is shown
in Table 2. When mole-rat scent
was present, mice behaved more similarly to their sibs that did not encounter
predator scent than to those that did. Thus males that encountered mole-rat
scent behaved similarly to their sibs that did not encounter predator scent in
terms of time spent beyond the tunnel, marking rate on the clean block or time
spent on the competitor-marked block (in either high- or low-marking
groups).
|
In contrast, males that encountered mole-rat scent behaved differently, in several ways, to their sibs that encountered predator scent (at least in high-markers). Thus high-markers that encountered mole-rat scent spent more time on the experimental side and on the competitor-marked block, and they scent-marked at a higher rate on the clean block, than their sibs that were exposed to predator odor. Latency data for high and low-markers were pooled to satisfy sample size requirements of Wilcoxon tests. Males exposed to mole-rat scent were slower to pass through the tunnel (T = 8, n = 10 pairs, p < 0.05) and to visit the marked block (T = 8, n = 10, p <.05) than their sibs that did not encounter predator scent. These effects were the only indications that mole-rat scent influenced subject behavior. Nonetheless, these delays in passing through the tunnel and reaching the marked block were far smaller when compared with sibs exposed to predator scent (tunnel: T = 0, n = 10 pairs, p <.01; block, T = 1, n1 = 10, n2 = 9, p <.01).
Taken together, these results suggest that mole-rat scent was perceived as being far less threatening than the predator scent, despite the two scents being equally unfamiliar. Furthermore, there was no indication that mole-rat scent affected behavior in any way other than initial investigation of the odor; notably (unlike the predator odor), it had no effect on subsequent levels of scent-marking.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Several studies have demonstrated effects on foraging (Epple et al., 1993
;
Jedrzejewski et al., 1993;
Kotler et al., 1992), spacing
(Abramsky et al., 1996),
avoidance (e.g., Gorman, 1984;
Jedrzejewski et al., 1993;
Robinson, 1990;
Stoddart, 1976) and defensive
(Hendrie et al., 1998)
behavior in rodents exposed to the scent of a predator, but none have
investigated effects on scent-marking. Here we have shown that animals that
differ in investment in scent-marking, a conspicuous status signaling
behavior, also differ in their response to a perceived increase in predation
risk.
Comparisons between the response of mice to the presence or absence of
predator scent and the unfamiliar but benign scent of the naked mole-rat
showed that subjects differentiated between the two unfamiliar scents and
responded much more strongly to the predator scent. Similar discrimination has
been found in meadow voles (Microtus pennsylvanicus) in tests between
odor of weasels (Mustela erminea) and guinea pigs (Cavia
porcellus; Parsons and
Bondrup-Nielsen, 1995), and in bank voles (Clethrionomys
glareolus) between several mustelid predators and rabbits
(Oryctolagus cuniculus;
Jedrzejewski et al., 1993).
There would therefore appear to be some qualitative component of the ferret
scent that produces the significant aversive response, rather than a reaction
to unfamiliar smells per se. Indeed, this type of fine-tuned identification of
predators by rodents is not restricted to their scent: several species exhibit
particular responses to owl calls but not to the same calls played backwards
(Hendrie et al., 1998). These
effects must point to innate recognition of the predator odor or its specific
properties, since our subjects had not been exposed to this scent before (nor,
presumably, had their ancestors of many previous generations). A similar,
apparently innate, response has recently been reported in water voles,
Arvicola terrestris, to the odor of American mink, Mustela
vison (Baretto and Macdonald,
1999). While such apparently adaptive responses might have
disappeared during domestication, this process cannot have produced them.
Both high- and low-frequency scent-marking males were slower to respond to a simulated territorial intrusion by a competitor (approaching the competitor's scent marks) in the presence of predator scent than siblings that were not exposed to predator scent. However, only high-marking males under predation risk reduced the amount of time they spent overmarking the intruder's scent. This appears to support our second hypothesis for the response of high-frequency signaling males: that the potential costs of predation cause these males to reduce their rates of scent-marking.
The hypothesis that high-markers would take greater risks in signaling than
low-markers is not supported, as low-markers in fact increase their signaling
investment in the presence of predation risk when high-markers reduce it.
Aside from scent-marking itself, however, high-markers were generally more apt
to take risks than low-markers, being much quicker to pass through the tunnel
and approach the competitor's marks for the first time and spending more time
in the vicinity of this simulated intruder's scent (in the experimental
section). They thus entered more readily and spent more time in areas of the
cage that contained predator scent. Although in our experimental design there
was no possibility of the individuals meeting, each of these behaviors would
otherwise tend to increase the chances of the subject encountering and
challenging the intruder. High-marking males therefore continue to behave as
we would expect of territorial or dominant individuals of high competitive
ability, with the exception of behavior specifically associated with
counter-marking. These findings suggest that there is a unique danger inherent
to scent-marking for males which typically mark at high frequencies. If
predators do make use of variable densities of scent marks as a cue to likely
prey density and foraging success rates (cf.
Viitala et al., 1995), then
high-marking individuals will become more vulnerable to predation than those
which mark their territory or range at lower frequencies. The benefits to
reduction in marking rate are therefore relatively larger to high-than
low-frequency markers and our observation that high-but not low-markers reduce
their level of counter-marking supports this idea.
Our finding that low-marking males scent-marked at higher rates when they
encountered predator odor was unexpected, but intriguing. More information is
needed about the social strategies of low-markers to fully explain why they
did not decrease their levels of investment in a way similar to that observed
in high-markers. However, one explanation is that if scent-marking by
high-quality animals helps to maintain dominance, but this effect is
diminished under predation risk because they reduce signaling rates, then
predation risk may cause low-markers to opportunistically increase
scent-marking investment. This would enhance their relative competitiveness,
and the likelihood of their achieving higher status, at a critical time. In
the current context, this would be adaptive if there is competition for refuge
sites, as is known to be the case in wild rodents, in which dominants have
more ready access to safer habitat sub-types
(Koivunen et al., 1998; see
also Gray and Hurst, 1997).
Low-markers may thus increase their signaling effort in order to have a better
chance of securing such access. Alternatively, they may adopt a different
mating strategy from high-markers which makes them less susceptible to
predation, perhaps varying in the extent to which they patrol territories,
challenge competitors or search widely for mates.
This study has shown that males of different quality vary in their response to predator odors and that, as in acoustic signaling systems, predation risk may cause counter-selection on olfactory signaling behavior. Our observations, in a closely-controlled model system, suggest useful avenues for investigation in both captive and wild contexts. Further study is required to investigate the consequences of differential scent-marking responses on male fitness, particularly in view of the possibly different mating strategies of high and low-markers. In addition, we need to determine the time period over which the described effects are sustained. It may be that they are essentially short-term responses to the threat of elevated predation risk. On the other hand, the reduction in the large difference between marking rates of high- and low-quality individuals might represent a predator-induced shift in individual behavior, towards more continuous levels of marking which may have been expressed ancestrally in a more natural, predator-rich environment. In either case, the results demonstrate the potential of predation risk as a powerful agent of counter-selection on olfactory signaling behavior.
|
ACKNOWLEDGEMENTS |
|---|
We are grateful to Chris Carbone for comments on the manuscript and to Mandy Gordon, Jenny Litten, Jake Rozowski, and Carol Williams for their help in mouse husbandry. The work was funded by the Leverhulme Trust and was carried out under license from the UK Home Office and within guidelines laid out under the Animals (Scientific Procedures) Act 1986, the Association for the Study of Animal Behavior and the Institute of Zoology's Ethics Committee.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Abramsky Z, Strauss E, Subach A, Kotler BP, Ricchman A, 1996. The effect of barn owls (Tyto alba) on the activity and microhabitat selection of Gerbillus allenbyi and Gerbillus pyramidum. Oecologia 105: 313-319.
Bailey WJ, Haythornthwaite S, 1998. Risks of calling by the field cricket Telegryllus oceanicus; potential predation by Australian longeared bats. J Zool 244: 505-513.
Baretto GR, Macdonald DW, 1999. The response of water voles, Arvicola terrestris, to the odors of predators. Anim Behav 57: 1107-1112.[Web of Science][Medline]
Belwood J, Morris GK, 1997. Bat predation and its influence on calling behavior in neotropical katydids. Science 238: 64-67.
Brain PF, Kamal KBH, 1989. Effects of prior social experiences on individual aggressiveness in laboratory rodents. Rassegna Psicol 6: 37-43.
Bronson FH, Marsden HM, 1973. The preputial gland as an indicator of social dominance in male mice. Behav Biol 9: 625-628.[Web of Science][Medline]
Brown RE, Macdonald DW, 1985. Social odors in mammals. Oxford: Clarendon Press.
Burk T, 1982. Evolutionary significance of predation on sexually signaling males. Fla Entomol 65: 90-104.
Cade WH, 1975. Acoustically orienting parasitoids: fly phonotaxis to cricket song. Science 190: 1312-1313.
Collins SA, Gosling LM, Hudson J, Cowan D, 1997. Does behavior after weaning affect the dominance status of adult male mice (Mus musculus)? Behavior 134: 989-1002.
Corridi P, Chiarotti F, Bigi S, Alleva E, 1993. Familiarity with conspecific odor and isolation-induced aggressive behavior in male mice (Mus domesticus). J Comp Psychol 107: 328-335.[Web of Science][Medline]
Desjardins C, Maruniak JA, Bronson FH, 1973. Social
rank in house mice: differentiation revealed by ultraviolet visualization of
urinary marking patterns. Science 182:
939-941.
Endler JA, 1992. Signals, signal conditions and the direction of evolution. Am Nat 139: 125-153.
Epple G, Mason JR, Nolte DL, Campbell DL, 1993. Effects of predator odors on feeding in the mountain beaver (Aplodontia rufa). J Mammal 74: 715-722.
Ferkin MH, 1999. Meadow voles (Microtus pennsylvanicus, Arvicolidae) over-mark and adjacent-mark the scent marks of same-sex conspecifics. Ethology 105: 825-837.
Gorman ML, 1984. The response of prey to stoat (Mustela erminea) scent. J Zool 202: 419-423.
Gosling LM, 1981. Demarkation in a gerenuk territory: an economic approach. Z Tierpsychol 56: 305-322.
Gosling LM, 1982. A reassessment of the function of scent marking in territories. Z Tierpsychol 60: 89-118.[Web of Science]
Gosling LM, 1986. Economic consequences of scent marking in mammalian territoriality. In: Chemical signals in vertebrates, vol. 4 (Duvall D, Müller-Schwarze D, Silverstein RM, eds). New York: Plenum Press; 385-395.
Gosling LM, 1990. Scent marking by resource holders: alternative mechanisms for advertising the costs of competition. In: Chemical signals in vertebrates, vol. 5 (Macdonald DW, Natynczuk S, Müller-Schwarze D, eds). Oxford: Oxford University Press; 315-328.
Gosling LM, Atkinson NW, Collins SA, Roberts RJ, Walters RL, 1996a. Avoidance of scent-marked areas depends on the intruder's body size. Behavior 133: 491-502.
Gosling LM, Atkinson NW, Dunn S, Collins SA, 1996b. The response of subordinate male mice to scent marks varies in relation to their own competitive ability. Anim Behav 52: 1185-1191.
Gosling LM, McKay HV, 1990. Competitor assessment by scent matching: an experimental test. Behav Ecol Sociobiol 26: 415-420.
Gosling LM, Roberts SC, 2001. Scent-marking by male mammals: cheat-proof signals to competitors and mates. Adv Stud Behav 30: 169-217.
Gosling LM, Roberts SC, Thornton EA, Andrew MJ, 2000. Life history costs of olfactory status signalling in mice. Behav Ecol Sociobiol 48: 328-332.
Grafen A, 1990. Biological signals as handicaps. J Theor Biol 144: 517-546.[Web of Science][Medline]
Gray SJ, Hurst JL, 1997. Behavioral mechanisms underlying the spatial dispersion of commensal Mus domesticus and grassland Mus spretus. Anim Behav 53: 511-524.
Hendrie CA, Weiss SM, Eilam D, 1998. Behavioral responses of wild rodents to the calls of an owl: a comparative study. J Zool 245: 439-446.
Huntingford FA, 1984. Some ethical issues raised by studies of predation and aggression. Anim Behav 32: 210-215.
Hurst JL, 1990. Urine marking in populations of wild house mice Mus domesticus Rutty. I. Communication between males. Anim Behav 40: 209-222.
Hurst JL, Hayden L, Kingston M, Luck R, Sorensen K, 1994. Response of the aboriginal house mouse Mus spretus Lataste to tunnels bearing the odors of conspecifics. Anim Behav 48: 1219-1229.
Jedrzejewski W, Rychlik L, Jedrzejewska B, 1993. Responses of bank voles to seven species of predators: experimental data and their relevance to natural predator-vole relationships. Oikos 68: 251-257.
Johnson RP, 1973. Scent marking in mammals. Anim Behav 21: 521-535.
Johnston RE, Sorokin ES, Ferkin MH, 1997a. Female voles discriminate males' over-marks and prefer top-scent males. Anim Behav 54: 679-690.[Web of Science][Medline]
Johnston RE, Sorokin ES, Ferkin MH, 1997b. Scent counter-marking by male meadow voles: females prefer the top-scent male. Ethology 103: 443-453.
Koivula M, Viitala J, 1999. Rough-legged buzzards use vole scent marks to assess hunting areas. J Avian Biol 30: 329-332.
Koivunen V, Korpimäki E, Hakkarainen
H, 1998. Refuge sites of voles under owl predation risk: priority
of dominant individuals? Behav Ecol 9:
261-266.
Kotler BP, Blaustein L, Brown JS, 1992. Predator facilitation—the combined effects of snakes and owls on the foraging behavior of gerbils. Annal Zool Fenn 29: 199-206.
McGregor PK, 1993. Signalling in territorial systems: a context for individual identification, ranging and eavesdropping. Phil Trans R Soc Lond B 340: 237-244.
Mendl M, Paul ER, 1990. Litter composition affects parental care, offspring growth and the development of aggressive behavior in wild house mice. Behavior 116: 90-108.
Parmigiani S, Mainardi D, Pasquali A, 1981. A comparison of aggressiveness in dominant subordinate and isolated house mice. In: The biology of aggression (Brain PF, Benton D, eds). Alphen aan den Rijn, Netherlands: Sijthoff & Noordhoff; 563-570.
Parsons GJ, Bondrup-Nielsen S, 1995. Experimental analysis of behavior of meadow voles (Microtus pennsylvanicus) to odors of the short-tailed weasel (Mustela erminea). Ecoscience 3: 63-69.
Rich TJ, Hurst JL, 1999. The competing countermarks hypothesis: reliable assessment of competitive ability by potential mates. Anim Behav 58: 1027-1037.[Web of Science][Medline]
Roberts SC, 1998. Behavioral responses to scent marks of increasing age in klipspringer Oreotragus oreotragus. Ethology 104: 585-592.
Roberts SC, Lowen C, 1997. Optimal patterns of scent marks in klipspringer (Oreotragus oreotragus) territories. J Zool 243: 565-578.
Robinson I, 1990. The effect of mink odor on rabbits and small mammals. In: Chemical signals in vertebrates, vol. 5 (MacDonald DW, Müller-Schwarze D, Natynczuk SE, eds). New York: Oxford University Press; 566-572.
Sakaluk SK, Belwood JJ, 1984. Gecko phonotaxis to cricket calling song: a case of satellite predation. Anim Behav 32: 659-662.
Sokal RR, Rohlf FJ, 1995. Biometry, 3rd ed. New York: Freeman.
Stoddard PK, 1999. Predation enhances complexity in the evolution of electric fish signals. Nature 400: 254-256.[Medline]
Stoddart DM, 1976. Effect of the odor of weasels (Mustela nivalis L.) on trapped samples of their prey. Oecologia 22: 439-441.
Trivers RL, 1972. Parental investment and sexual selection. In: Sexual Selection and the Descent of Man, 1871-1971 (Campbell B, ed). London: Heinemann Press; 136-179.
Tuttle MD, Ryan MJ, 1981. Bat predation and the
evolution of frog vocalisations in the neotropics. Science
214: 677-678.
Viitala J, Korpimäki E, Palokangas P, Koivula M, 1995. Attraction of kestrels to vole scent marks visible in ultraviolet light. Nature 373: 425-427.
Williams GC, 1966. Sex and evolution. Princeton, New Jersey: Princeton University Press.
Zahavi A, 1975. Mate selection: a selection for a handicap. J Theor Biol 53: 205-214.[Web of Science][Medline]
Zuk M, Simmons L, Cupp L, 1993. Calling characteristics of parasitized and unparasitized populations of the field cricket Teleogryllus oceanicus. Behav Ecol Sociobiol 33: 339-343.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  F. Rosell and J. Sanda Potential risks of olfactory signaling: the effect of predators on scent marking by beavers Behav. Ecol., November 1, 2006; 17(6): 897 - 904. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. O. Wolff Scent marking by voles in response to predation risk: a field-laboratory validation Behav. Ecol., March 1, 2004; 15(2): 286 - 289. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. M. Zala, W. K. Potts, and D. J. Penn Scent-marking displays provide honest signals of health and infection Behav. Ecol., March 1, 2004; 15(2): 338 - 344. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||