Behavioral Ecology Vol. 12 No. 5: 584-589
© 2001 International Society for Behavioral Ecology
Conditional allogrooming in the herb-field mouse
Wildlife Conservation Research Unit, Department of Zoology, University of
Oxford, South Parks Road, Oxford OX1 3PS, UK, and Biodiversity Research Group,
Department of Zoology, Charles University,
Vini
ná 7,
Prague 2, CZ 128 44, The Czech Republic
Wildlife Conservation Research Unit, Department of Zoology, University of
Oxford, South Parks Road, Oxford OX1 3PS, UK, and Biodiversity Research Group,
Department of Zoology, Charles University,
Vininá 7,
Prague 2, CZ 128 44, The Czech Republic
Address correspondence to P. Stopka, who is now at Biodiversity Research
Group, Department of Zoology, Charles University,
Vininá 7, 128
44, Prague 2, The Czech Republic. E-mail:
pstopka{at}natur.cuni.cz
.
Received 7 December 1999; revised 3 October 2000; accepted 16 November 2000.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Among members of the family Muridae, the herb-field mouse, Apodemus microps, is unique in that aggression is almost entirely lacking. This species, therefore, is a model organism for experimental studies of social behavior without the confounding influence of aggression. We used video surveillance cameras to assess the importance of self-grooming and allogrooming in the social life of this species. Detailed analysis of individual behavioral sequences using Markov chain methods revealed that self-grooming is a relatively stereotypic, sex-independent activity usually lasting about 8 s. Allogrooming is conditional in the herb-field mouse, because it takes the form of a reciprocal strategy, with the differences between nonmatching bouts varying according to whether the initiator of allogrooming is male or female and whether both interactants are of the same or opposite sex. Our analysis revealed that the exchange of allogrooming bouts between individuals of the same sex is reciprocal, but that males allow females to "defect" more often than vice versa, and males groomed females for longer than predicted by the distribution of individual self-grooming bouts. In those species where the demand for mating by males is far greater than that offered by females, in other words where females may select mates, asymmetry of allogrooming may provide a mechanism for females to test the "suitability" of males for mating. It may also provide the means for males to stimulate females before mating. Furthermore, allogrooming was the only sex-dependent behavior of the several tested in our experiment. As such, we suggest that allogrooming is the predominant premating mechanism in this species.
Key words: allogrooming, Apodemus microps, cooperation, programmed grooming, reciprocity.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The exchange of reciprocally advantageous acts has been documented in many mammals and was often interpreted as altruism between unrelated individuals (Trivers, 1971
). Recently,
various functional strategies have been shown to account for observed
deviations from reciprocity of the individuals involved
(Nowak and Sigmund, 1993).
They also show, however, that the development of social relationships may be
explained through increasing investment in the course of ongoing cooperative
encounters, as in the case of the "raise-the-stakes" strategy (see
Roberts and Sherratt, 1998).
When the "payoff" largely depends on constraints, such as the
availability of food or mates, behaviors are often exchanged for other
behaviors or commodities, such as support, as occurs in chimpanzees
(Hemelrijk, 1991), food in
baboons (Barrett et al., 1999),
or mating in wood mice (Stopka and
Macdonald, 1999). Further, it has been evidenced in numerous
studies that a "biological market" is the underlying mechanism
explaining this asymmetry
(Noë and
Hammerstein, 1994;
Noë et al.,
1991). In comparison with classical models of cooperation based on
the prisoner's dilemma the model of "biological markets" is based
on a fluctuating payoff, largely determined by market forces. The payoff,
therefore, is not stable but instead acquires a highly dynamic property.
Investments such as provision of grooming are, therefore, exchanged for the
same or other social services in an individually specific fashion dependent on
the supply and demand for a particular commodity.
Studying the properties of reciprocal behavior, however, requires
consideration of the costs involved in order to distinguish cooperation from
no-cost cooperation, that is, by-product mutualism (see
Dugatkin, 1997 for review). At
the level of behavioral sequences during interactions it has been suggested
that a cost is always involved when changing from one activity to another
(Larkin and McFarland, 1978).
The cost of changing between behavior patterns that cannot be performed
simultaneously may influence the timing of decisions and is, therefore,
responsible for changing the structure of behavioral sequences
(McFarland, 1971). In wild
animals it has been shown that time is a constraint which shapes the
"decision structure." For example, baboons switch from other
activities to feeding more often when in lower quality habitats
(Dunbar, 1992). Furthermore,
when time is crucial, as in instances of food shortage, the cost of
allogrooming is even higher, and individual baboons did not invest time into
social activities such as allogrooming
(Dunbar, 1992). This may lead
to increased parasite infestation. It has been also demonstrated that in
domestic and wild mammals such as moose
(Mooring and Samuel, 1998)
tick infestation may lead to a decrease in survival probability during harsh
environmental conditions. In the yellow-bellied marmot increased ectoparasite
loads caused a decrease in growth rate, low overwinter survival, and reduced
reproduction (van Vuren,
1996). The cost of providing allogrooming may increase over time
as proposed by the vigilance principle
(Ruxton and Roberts, 1999)
which states that an individual's vigilance against predators is decreased
during interactions mainly because vigilance sequences are infrequent or
short, due to allogrooming (Maestripieri,
1993; Mooring and Hart,
1995). Also, the groomer may suffer water and electrolyte (saliva)
loss and tooth wear (Mooring and Hart,
1995). Such costs may also accrue in animals where parasite
infestation is low, as in beavers, where grooming is primarily important in
maintaining waterproofing properties of fur
(Patenaude and Bovet, 1984).
Neither can one exclude the possibility that allogrooming is beneficial for
other reasons, such as stimulating blood circulation or stimulating the
release of beta endorphins (Keverne et
al., 1989) which cause relaxation, as has been documented in
horses where grooming in certain places reduced heart rate
(Feh and
Mazières, 1993). In the wood mouse
(A. sylvaticus), males of which are more aggressive to others of
their own sex than are females, the demand for matings by males is far larger
than the number of matings females can offer. This often results in
multiplysired litters (Baker et al.,
1999) and agonistic interactions among males
(Montgomery, 1978).
Allogrooming in this species takes the form of bargaining, because males have
to groom females more often in order to obtain information about the
reproductive state of females (Stopka and
Macdonald, 1999) and/or access to mating
(Stopka and Macdonald, 1998).
Furthermore, allogrooming was "demanded" by a female before a male
was allowed to obtain information about her reproductive state via anogenital
sniffing. Mating also consistently followed a long bout of allogrooming
provided by males. The "demand" for grooming was enforced by mild
agonistic behavior from the female if it was not provided. Such a form of
behavioral interchange is compatible with models of biological markets
(Noë et al.,
1991;
Noë and
Hammerstein, 1994). In this article we concentrated on social
grooming in a poorly studied muroid rodent, the herb-field mouse, A.
microps. This species is a common member of the small-mammal communities
in Central Europe, living in arable—often ploughed—fields
(Pelikán,
1970) and was recently described as the least aggressive species
of the genus Apodemus
(Suchomelová
and Frynta, 2000).
As the social biology of this species is not yet well known the main aim
was to study the geometry of social relationships by explaining deviations the
reciprocity of grooming. Here the term reciprocity is understood in accordance
with the definitions of Hemelrijk
(1990): "absolute"
means exchanged bouts are approximately matching; "relative" means
grooming given versus that received may be asymmetrical, but the proportion of
grooming received relatively matches the grooming given out to all other
members of the social group. Reciprocity of allogrooming has previously been
studied in impalas where individuals alternate in allogrooming
(Hart and Hart, 1992) and in
badgers (Meles meles) where individuals groom mutually using a
generous tit-for-tat-like rule set (i.e., when one groomer stops—the
other one stops too until one of them resumes allogrooming), which ensures
that similar durations of grooming are given and received
(Stewart, 1997). In the
monogamous white-handed gibbon (Hylobates lar), males provide less
paternal investment but more grooming than they receive back from the female.
By contrast, in the closely related and also monogamous siamang (Hylobates
syndactylus) male—female allogrooming is reciprocal and symmetrical
with males investing more into parental care
(Palombit, 1996). In the
herb-field mouse, however, individuals do not alternate during allogrooming
(i.e., grooming is not exchanged instantaneously) because individuals
intermittently switch between different behaviors that cannot be performed at
the same time. This raises the fundamental question: do individual mice
reciprocate bouts of allogrooming over a longer time interval (i.e.,
consecutively)? This leads to further questions: If it does occur, does such
an organization of sequences reveal a stable pattern allowing for the building
of relationships between particular mice? Can the pattern of allogrooming be
explained on the basis of the programmed grooming hypothesis (Mooring and
Hart, 1998) as in the case of self-grooming? As far as we are aware, this is
the first study of cooperation in this species.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Subjects and procedure
Our experiments strictly follow the requirements for the use of animals in research according to the current law in the Czech Republic. At all stages of our work with mice we rigorously observed the ethical requirements as laid out, for example, in Dawkins and Gosling (1992
).
Ten males and ten females of A. microps—caught in two
localities in the Southern parts of Moravia (the localities of Drnholec and
Jemeni
t
)—were
used in this experiment. To minimize the possible confounding effect of
familiarity or relatedness between the four individuals we used established
social groups, capturing individuals at two different sites and at places
remote from each other. After capture all individuals were weighed and
immediately transferred into plastic cages (39 x 23 x 23 cm) where
they were kept individually under a 12:12 LD cycle for three weeks prior to
observation to further dilute any possible effects of familiarity as mentioned
above. When caught (and tested) all individuals were sub-adults weighing 14.4
± 2.1 g and none of the females was pregnant. One day before video
observation started all mice were individually marked with black Rhodol-D (a
nontoxic dye) that has been used in similar studies (see
Stewart et al., 1997). All
individual and social activities were observed under infrared illumination
which is supposed to be outside of normal mammalian vision
(Lythgoe, 1979).
We studied grooming using two approaches. Rhythmicity and the duration of
self-grooming bouts were estimated from individual activities recorded in a
neutral arena (plastic cages—39 x 23 x 23 cm) using the open
field test (e.g., Gurnell,
1977). In this procedure every individual was recorded in random
order to exclude time and sequence effects. Recording for 10 min per mouse
took place between 1700 h and 2100 h and after each observation all animals
were transferred back to their cages while the observation cage was cleaned
with ethanol before the next trial. Allogrooming was studied in larger glass
cages (57 x 48 x 58 cm) also using video surveillance cameras,
excluding the possibility of observer bias. Two males and two females were
placed in the cage for 14 h beginning at 1800 h. These five groups were
consecutively recorded using continuous recording on a time-lapse video
recorder. The video camera's field of view covered the entire cage. Therefore,
every behavior recorded had a beginning and an end
(Bressers et al., 1991) and was
defined on the basis of its function (grooming, scanning, mobile, immobile,
etc.) and whether it was a part of interaction or an individual behavior. A
bout was defined as the repetitive action of the same behavior (e.g., an
allogrooming bout was the sum of uninterrupted states such as grooming head,
neck, flanks, back). To reduce the experimenter bias all behaviors were
analyzed frame by frame from videotapes with the program, Noldus VideoPro v.
4.0 (Noldus, 1997). Behavioral
matrices and sociomatrices were constructed and analyzed by the program The
Observer (Noldus, 1994).
The open field test
Null hypotheses of this test were: (1) distributions of durations of
self-grooming bouts between males and females are identical and (2) there is
no significant difference in the total time spent self-grooming per time unit
between males and females. If these hypotheses were not falsified we could
assume that males and females (without fleas and ticks) require the same
amount of "programmed" grooming per time unit. This step had
further importance in clarifying that any possible lack of reciprocity was not
due simply to general differences in grooming requirements by opposite sexes
as predicted by the "vigilance principle." This principle states
that males will groom less than females because they need to remain vigilant
against rival males and/or for receptive females and cannot, therefore, invest
so much time in other activities, including self-grooming
(Mooring et al., 1996).
Further differences may be due to testosterone which may have a downregulating
effect on grooming rates in males (Mooring
et al., 1998).
All behaviors were tested for sex-related differences in the distribution
of bout lengths from both the individual activities in the neutral arena and
in the enclosure. General differences between the distributions of activities
were tested with a General Linear Model procedure using an F test and
the means and total bout length were compared using a T test with
program Systat v. 5.01. All individual activities were also tested for
Markovian properties. Distributions and extinction rates were estimated from
log-survivor plots and formally tested for deviations from exponentially using
a Kolmogorov-Smirnov test (Haccou and
Meelis, 1992). Individual probabilities were then combined using
Fisher's omnibus combination procedure to obtain overall results on the
deviations from exponentiality. To clarify that self-grooming bouts are
independent of other behaviors we also tested the sequential properties of all
behaviors using the 
2 test for first-against second-order
dependency in the sequences of acts
(Haccou and Meelis, 1992).
Generally speaking, when the distribution of bout lengths for given behaviors
is approximately exponential and there is at most a first order dependency,
then these behavioral elements can be considered as independent of each
other—they have a Markovian property
(Haccou and Meelis, 1992).
Social interactions
In our study of individual activities, the Continuous Time Markov Chain was
the best model for the generalization of self-grooming. This behavior was,
therefore, analyzed in terms of the predictability over time of revealing
information about the "decision structure" of this behavior
(Dawkins and Dawkins, 1974).
Social grooming, however, required a more deterministic causal model because
allogrooming bouts were clustered in several consecutive grooming sessions.
Stationarity—which is one of the assumptions for Markov models—was
not, therefore, satisfied and this made such models unfeasible.
Here we tested the null hypothesis that allogrooming is reciprocated and,
therefore, the total time spent allogrooming per interaction between two given
individuals was not expected to be significantly different. We employed the
row-wise matrix correlation test to find the relative correlation between
actor and receiver matrices and the Mantel permutation test for testing more
specific levels of association between sociomatrices respecting directionality
of behaviors (see Hemelrijk,
1990) using program Matman v. 3.0
(de Vries et al., 1993).
General differences between distributions of different bouts were tested
visually using log-survivor plots and formally using the F test or
the Kolmogorov-Smirnov test.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
How important is grooming for individuals of different sex?
No individual had fleas or ticks before or after the experiment. This made it possible to assume that only the internal mechanisms or at most the current quality of the fur is a stimulus for self-grooming. Using the open field test, we found that there is no evidence that the distributions of self-grooming bouts are different between sexes (F(1,18) = 0.1717; p =.6835; see Figure 1) and that the total time spent self-grooming per time unit does not differ between sexes (t test 2-tailed; t = 0.306, df = 18; p =.763). This lack of differences between males and females in the open field test was also confirmed by analyzing the time spent self-grooming per time unit from the observations in the enclosure (t test 2-tailed; t = 1.063; DF = 18; p =.309). Therefore, we assumed that individuals of both sexes require an equal amount of self-grooming.
|
Self-grooming is often considered a relatively stereotypic activity, where
the direction of this behavior is hierarchically organized
(Fentress and Stilwell, 1973).
In this study, the distribution of the self-grooming bouts was exponential, as
were the distributions of gaps between subsequent bouts. Self-grooming in the
herb-field mouse, therefore, can be described as a stochastic process where
bouts of any length given in the log-survivor plot (see
Figure 1) are equally likely to
happen—that is, bout lengths do not deviate from exponentiality
(Fisher's omnibus procedure, f = 51.817, df = 2*
N = 40, p =.1). The time of initiation of such bouts is also
random, as suggested by the exponential distribution of the gaps between
consecutive bouts (Fisher's omnibus procedure, f = 50.824, df = 40,
p = 0.11). This is in agreement with the prediction of the programmed
grooming hypothesis, which states that an internal clock or endogenous
generator evokes a bout of grooming (even without ectoparasites as in this
study) at more or less regular intervals in order to remove parasites before
they attach and blood-feed (e.g., Hart et
al., 1992; Mooring et al.,
1998).
The analysis of allogrooming, however, revealed that there is a difference in the length of allogrooming given by individuals of different sexes. As presented in Figure 2, the extinction rates of allogrooming bouts provided by females are steeper than those provided by males. This difference is significant (Kolmogorov-Smirnov test, two-sided probability, KS = 0.248, p =.001). This means that males provide other males as well as females with significantly longer allogrooming bouts than females. There was, however, no evidence that the distributions of bouts provided by females to other females and to males are different (KS = 0.130, p =.806) nor are the bouts provided by males to other males or females (KS = 0.135, p =.472).
|
Is allogrooming reciprocal?
Individuals of both sexes were involved in two common types of
allogrooming. The first type, which was extremely rare, was simultaneous
allogrooming where two individuals were close to each other and groomed those
parts of each other that were accessible from a particular angle or position.
The most common type, however, was the one where two individuals allogroom
without instantaneous reciprocation (i.e., individuals do not alternate).
During the course of such interactions an individual would often be groomed in
those parts of the body that are not accessible during individual
self-grooming bouts (e.g., head, neck). While being groomed in these areas, an
individual is often disabled from any other action including the reciprocation
of allogrooming. Based on predictions of reciprocal altruism, this type of
allogrooming should also be reciprocal. Therefore, we tested the null
hypothesis that allogrooming is reciprocated over total time of 14 h of
observation. Furthermore, we tested whether the rank of durations of
allogrooming provided to particular individuals correlates with the rank of
durations received from them. We employed the Mantel permutation test to find
out whether the actor matrices are significantly associated with the receiver
matrices (i.e., transposed version of the actor matrix). This test was
followed by a less specific alternative, the row-wise matrix correlation test
(i.e., rk), applied to test the relative reciprocity based on the
correlation between actor and receiver matrices (see
Hemelrijk, 1990). Indeed, the
rank of durations of allogrooming given was significantly positively
correlated with the rank of durations received from those individuals (Mantel
test, p <.01; row-wise correlation, p <.0002, using
5000 permutations) revealing a nonrandom pattern of allogrooming exchanged
between different individuals. To remove the effect of influence of reciprocal
allogrooming between individuals of the same sex we performed the same test
only for male—female interactions. Again, the rank of durations of
allogrooming given to particular females by individual males correlated
(positively) with the rank of durations received from them (Mantel test,
p <.001; row-wise correlation, p <.0002). This is
evidence that the grooming provided to particular females is relatively
reciprocated and nonrandom as concerns the identity of individuals. To provide
further evidence of the relationship between grooming and mating, we performed
a correlation analysis between matrices of grooming interactions between males
and females and the corresponding matrix of durations of naso-anogenital
contacts. There was a significant association and positive correlation between
both matrices (Mantel test, p = 0.022; row-wise correlation,
p =.0005) revealing that those males which groomed particular females
more often were also those which succeeded in attaining more anogenital
contacts from them (significance levels of statistical tests did not change
when the frequencies were used instead of the sum of durations). There was,
however, a genuine lack of correlation (p =.19) between the total
time spent interacting with any of the different females and attained
anogenital contacts, as well as a lack of correlation between naso-anogenital
contacts and allogrooming between individuals of the same sex. This means that
only those males which provided sufficient allogrooming were able to proceed
with the identification of the female reproductive state.
To further demonstrate the quantitative level of the differences between grooming given and received we have employed a simple regression analysis. As presented in Figure 3, the linear relationship between bouts given and received reliably indicates reciprocity when bouts are exchanged between individuals of the same sex (females, p =.003; males, p =.003). However, grooming between males and females was largely asymmetrical. Although allogrooming between given two individuals of the opposite sex is relatively reciprocal, the slope of the regression line in Figure 4 (a = 0.2618) indicates that the given bouts are not reciprocated equally (y = 0.2618x + 10.373; p = 0.0001; p = 0.00001 when the outlier in Figure 4 is included). Therefore, we tested whether the bout difference (obtained as the bout received, subtracted from the bout given) varies across individuals in respect of their sex. Sex had a significant effect on the variation of this difference across different individuals (Mann-Whitney statistics, Z = 213.5; p =.037). Moreover, males allowed other individuals larger differences between bouts given and received than females. As presented in Figure 4, this is most likely to be due to males "allowing" females not to reciprocate allogrooming bouts. Using a game theory idiom, males are more likely to allow females to "cheat" in the process of cooperation rather than vice versa. In male—male interactions this is because the exchanged bouts were longer and, therefore, it is likely that the error would increase with the bout length too.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Detailed descriptions of grooming sequences are extremely important if we are to understand their relevance in both the individual and the social life of mammals, and especially if we are to understand how this behavior is organized (Fentress and Stilwell, 1973
). Such descriptions can often be confounded when studied in
rodents where grooming is often linked to another function, such as
aggression, broadcasting scent signals to conspecifics
(Ferkin et al., 1996), or
mating behavior (Stopka and Macdonald,
1998). Grooming in mammals takes two forms, self-grooming and
allogrooming. Self-grooming in the herb-field mouse is a stochastic process
where each bout is unpredictably but regularly performed even without a
presence of major ectoparasites such as ticks and fleas and this is in
agreement with the programmed grooming hypothesis
(Hart et al., 1992;
Mooring and Samuel, 1998).
Allogrooming as a means of hygienic behaviors directed to those places which
are difficult to reach—such as the head and neck—is important to
individuals with high parasite loads. Because allogrooming in the herb-field
mouse is exchanged in consecutive interactions, we tested the hypothesis that
reciprocity operates over a longer time scale.
Reciprocity of allogrooming has traditionally been studied in species where
individuals alternate in allogrooming such as impalas
(Hart and Hart, 1992), badgers
(Stewart, 1997), or baboons
(Barrett at al., 1999;
Dunbar, 1992). We have studied
cooperation in a species that rapidly switches between different behavioral
categories. All interactions are usually short and, therefore, allogrooming is
not reciprocated instantaneously, or within a relatively short period of time.
The main result of this study, therefore, is that allogrooming is reciprocated
within longer time intervals between individuals of the same sex, that it is
reciprocal but asymmetrical among individuals of the opposite sex, and that
this pattern is significant and nonrandom as revealed by the analysis of
sociomatrices. It is clear from Figure
4 that males groom females more often than vice versa while
providing females with longer grooming bouts. The slope of regression lines in
Figure 3 and
Figure 4, as well as relevant
statistics used, indicate that the relationship is consistent. Grooming
between individuals of the same sex is exchanged for grooming as there is no
other commodity to offer. However, allogrooming between males and females is
asymmetrical as there can be future benefits in bargaining grooming for mating
access. This is similar to the situation found in the wood mouse where
grooming is provided by males to obtain information on female reproductive
status (Stopka and Macdonald,
1999) or mating (Stopka and
Macdonald, 1998). In the wood mouse, however, those males which
are most aggressive to other males are also those which provide the most
grooming to females (Stopka and Macdonald,
1999). In the herb-field mouse, aggression is almost entirely
absent; this may explain why males are involved in reciprocal allogrooming
while females do not demand grooming using agonistic responses as in the wood
mouse (Stopka and Macdonald,
1999). Moreover, even without the presence of major ectoparasites,
herb-field mice allogroom in this non-random pattern revealing that it may be
programmed, similar to the programmed self-grooming suggested by Hart et al.
(1992).
This result presents a new view on the social biology of this species. The
herb-field mouse is a rodent with no sexual dimorphism at a morphological
level and, as we have shown in this study, demands for self-grooming were
equal among sexes. Among other species of wood mice males are usually more
aggressive (Montgomery, 1978;
iháková
and Frynta, 1996) and this is often explained in terms of
territorial defense or of maintenance of reproductive efficiency. Aggression
in this species, however, was lacking in all established social groups and
grooming interactions, as well as anogenital stimulation between individuals
of the same sex, although bouts of allogrooming between same-sex individuals
were approximately equally as likely to occur as such interactions between
individuals of the opposite sex. This may well account for the lack of
aggression and perhaps lack of scent communication. Although subcaudal scent
glands are well developed in males of other more aggressive species of the
genus Apodemus, they are poorly developed in the herb-field mouse
(Stoddart, 1971). Furthermore,
because the mating system is not yet known in this species, it seems likely
that the allogrooming behavior in the herb-field mouse may serve several
social functions. First, allogrooming between individuals of the same sex
probably reflects a common mechanism under which costs are minimized to the
extent that the amount of time given approximately matches the length of time
for which grooming was received. In this species, however, allogrooming is
reciprocated in consecutive interactions over a longer period, but reciprocity
holds. Second, the asymmetry in allogrooming between sexes probably reflects a
general trend also found in other rodents (Stopka and Macdonald,
1998,
1999) where males provide
females with grooming more "generously" in terms of reciprocity,
to obtain information about their receptivity (estrous/anestrous) and/or to
establish mating consort.
|
ACKNOWLEDGEMENTS |
|---|
We gratefully acknowledge the helpful comments of Dr. M. Mooring and two anonymous referees. We also acknowledge the Ministry of Education, Youth, and Sport of the Czech Republic (grant no. VS 97102) for funding this work. In addition, we thank the Royal Society (London) who awarded the fellowship to P.S. for work in Oxford. We thank D.D.P. Johnson, D.W. Macdonald, and Joanna Summers for their editorial help and helpful discussions, P. Nová for providing us with the studied individuals from the locality of Je
menit
(South Moravia) and J. Polechová and B.
Bímová for
their help during trapping. In addition, we are very grateful to P. Johnson
for his statistical advice. |
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Baker RJ, Makova KD, Chesser RK, 1999. Microsatellites indicate a high frequency of multiple paternity in Apodemus (Rodentia). Mol Ecol 8: 107-111.[Medline]
Barrett L, Henzi SP, Weingrill T, Lycett JE, Hill LA, 1999. Market forces predict grooming reciprocity in female baboons. Proc R Soc Lond B 266: 665-670.
Bressers WMA, Meelis E, Haccou P, Kruk MR, 1991. When did it really start or stop: the impact of censored observations on the analysis of durations. Behav Proc 23: 1-20.
iháková
J, Frynta D, 1996. Intraspecific and interspecific behavioural
interactions in the Wood mouse (Apodemus sylvaticus) and the
Yellow-necked mouse (Apodemus flavicollis) in a neutral cage.
Folia Zool 45:
105-113.
Dawkins M, Dawkins R, 1974. Some descriptive and explanatory stochastic models of decision making. In: Motivational control systems analysis (McFarland DJ, ed). London: Academic Press; 119-168.
Dawkins MS, Gosling M, 1992. Ethics in research on animal behaviour. London: Academic Press.
de Vries H, Netto WJ, Hanegraaf LH, 1993. Matman: a program for the analysis of sociometric matrices and behavioural transition matrices. Behaviour 125: 157-75.
Dugatkin LA, 1997. Cooperation among animals. Oxford: Oxford University Press.
Dunbar RIM, 1992. Time: a hidden constraint on the behavioural ecology of baboons. Behav Ecol Sociobiol 31: 35-49.
Feh C, Mazières De J, 1993. Grooming at a preferred site reduces heart rate in horses. Anim Behav 46: 1191-1194.
Fentress JC, Stilwell FP, 1973. Grammar of a movement sequence in inbred mice. Nature 244: 52-53.[Medline]
Ferkin MH, Sorokin ES, Johnston RE, 1996. Selfgrooming as a sexually dimorphic communicative behaviour in meadow voles, Microtus pennsylvanicus. Anim Behav 51: 801-810.
Gurnell J, 1977. Neutral cage behavioural interaction in wild wood mice, Apodemus sylvaticus (Linné, 1758). Säugetierk Mitt 25: 57-66.
Haccou P, Meelis E, 1992. Statistical analysis of behavioural data (An approach based on time-structured models). Oxford: Oxford University Press.
Hart BL, Hart LA, 1992. Reciprocal allogrooming in impala, Aepyceros melampus. Anim Behav 44: 1073-1083.
Hart BL, Hart LA, Mooring MS, Olubayo R, 1992. Biological basis of grooming behaviour in antelope: the body size, vigilance and habitat principles. Anim Behav 44: 615-631.
Hemelrijk CK, 1990. Models of, and tests for, reciprocity, unidirectionality and other social interaction patterns at a group level. Anim Behav 39: 1013-1029.
Hemelrijk CK, Ek A, 1991. Reciprocity and interchange of grooming and `support' in captive chimpanzees. Anim Behav 41: 923-935.
Keverne EB, Martensz N, Tuite B, 1989. ß-endorphin concentrations in cerebrospinal fluid of monkeys are influenced by grooming relationships. Psychoneuroendocrinology 14: 155-161.[ISI][Medline]
Larkin S, McFarland D, 1978. The cost of changing from one activity to another. Anim Behav 26: 1237-1246.
Lythgoe JN, 1979. The ecology of vision. Oxford: Clarendon Press.
Macdonald DW, Apps PJ, Carr G, Kerby G, 1987. Social dynamics, nursing coalitions and infanticide among farm cats (Felis catus). Adv Ethol 28: 1-66.
Maestripieri D, 1993. Vigilance costs of allogrooming in Macaque mothers. Am Nat 141: 744-753.
McFarland DJ, 1971. Feedback mechanisms in animal behaviour. London: Academic Press.
Montgomery WI, 1978. Intra- and interspecific interactions of Apodemus sylvaticus (L.) and A. flavicollis (Melchoir) under laboratory conditions. Anim Behav 26: 1247-1254.
Mooring MS, Gavazzi AJ, Hart BL, 1998. Effects of castration on grooming in goats Physiol Behav 64: 707-713.[Medline]
Mooring MS, Hart BL, 1995. Costs of allogrooming in impala: distraction from vigilance. Anim Behav 49: 1414-1416.
Mooring MS, McKenzie AA, Hart BL, 1996. Role of sex and breeding status in grooming and total tick load of impala. Behav Ecol Sociobiol 39: 259-266.
Mooring MS, Samuel WM, 1998. The biological basis of grooming in moose: programmed versus stimulus-driven grooming. Anim Behav 56: 1561-1570.[ISI][Medline]
Noë R, Hammerstein P, 1994. Biological markets: supply and demand determine the effect of partner choice in cooperation, mutualism and mating. Behav Ecol Sociobiol 35: 1-11.
Noë R, van Schaik CP, van Hoof JARAM, 1991. The Market effect: an explanation for pay-off asymmetries among collaborating animals. Ethology 87: 97-118.
Noldus Information Technology, 1994. The observer, base package for Windows. (reference manual: version 3.0 edition) The Netherlands: Noldus, Wageningen.
Noldus Information Technology, 1997. The observer, support package for video analysis, Windows edition (reference manual: version 4.0) The Netherlands: Noldus, Wageningen.
Nowak M, Sigmund K, 1993. A strategy of win-stay, lose-shift that out-performs tit-for-tat in the prisoner's dilemma game. Nature 364: 56-58.[Medline]
Palombit RA, 1996. Pair bonds in monogamous apes: a comparison of the siamang Hylobates syndactylus and the white-handed gibbon Hylobates lar. Behaviour 133: 321-356.
Patenaude F, Bovet J, 1984. Selfgrooming in the North American beaver, Castor canadensis. Can J Zool 62: 1872-1878.
Pelikán J, 1970. Sex ratio in three Apodemus species. Folia Zool 19: 23-34.
Roberts G, Sherratt TN, 1998. Development of cooperative relationships through increasing investment. Nature 394: 175-179.[Medline]
Ruxton GD, Roberts G, 1999. Are vigilance sequences a consequence of intrinsic chaos or external changes? Anim Behav 57: 493-495.[Medline]
Stewart PD, 1997. The social behaviour of European badger (Meles meles) (PhD Thesis). Oxford: University of Oxford.
Stewart PD, Ellwood SA, Macdonald DW, 1997. Remote video-surveillance of wildlife—an introduction from experience with the European badger, Meles meles. Mammal Rev 27: 185-204.
Stoddart M, 1971. An examination of the caudal organ of Apodemus agrarius, A. flavicollis, A. microps, and A. sylvaticus, from Moravia. Folia Zool 21: 39-42.
Stopka P, Macdonald DW, 1998. Signal interchange during mating in the wood mouse (Apodemus sylvaticus): the concept of active and passive signalling. Behaviour 135: 231-249.
Stopka P, Macdonald DW, 1999. The market effect in the wood mouse, Apodemus sylvaticus: selling information on reproductive status. Ethology 105: 969-982.
Suchomelová E, Frynta D, 2000. Intraspecific behavioural interactions in Apodemus microps: a peaceful mouse? Acta Ther 45: 201-209.
Trivers RL, 1971. The evolution of reciprocal altruism. Q Rev Biol 46: 35-57.
van Vuren D, 1996. Ectoparasites, fitness, and social behaviour of yellow-bellied marmots. Ethology 102: 686-694.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  R. L. Chancellor and L. A. Isbell Female grooming markets in a population of gray-cheeked mangabeys (Lophocebus albigena) Behav. Ecol., September 26, 2008; (2008) arn117v1. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. D. P. Johnson, P. Stopka, and D. W. Macdonald Ideal flea constraints on group living: unwanted public goods and the emergence of cooperation Behav. Ecol., January 1, 2004; 15(1): 181 - 186. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||