Behavioral Ecology Vol. 11 No. 3: 288-293
© 2000 International Society for Behavioral Ecology
The power of shell rapping influences rates of eviction in hermit crabs
School of Biology and Biochemistry, The Queen's University of Belfast, Medical Biology Centre, 97 Lisburn Road, Belfast BT9 7BL, UK
Address correspondence to M. Briffa. E-mail m.briffa{at}qub.ac.uk .
Received 6 June 1999; revised 6 September 1999; accepted 15 September 1999.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Hermit crabs fight for ownership of shells, and shell exchange may occur after a period of shell rapping, involving the initiating or attacking crab bringing its shell rapidly and repeatedly into contact with the shell of the noninitiator or defender, in a series of bouts. The temporal pattern of rapping contains information about the motivation and/or relative resource holding potential (RHP) of the initiator and acts as a repeated signal of stamina. Here we investigated the role of the force with which the rapping is performed and how this is related to the temporal pattern of rapping by rubberizing the external surface of shells. Initiators that are prevented from rapping with their usual level of force persist with the activity for longer over the whole encounter but use fewer raps per bout and are less likely to effect an exchange than those supplied with control shells. The fact that the force of rapping affects the likelihood of a crab being victorious suggests that either the force of rapping contains information about motivation or RHP or that force directly affects noninitiators, reducing their ability to maintain an adequate grip on their shells. The data suggest that shell rapping is an agonistic signal rather than one that provides information useful to the noninitiator, as has been suggested by the negotiation model of shell exchange.
Key words: aggression, agonistic behavior, communication, hermit-crabs, signal.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Studies of repeated signals that are used to advertise quality have been conducted on a range of animals such as hermit crabs (Briffa et al., 1998
), various
anurans (Lopez et al., 1988;
Wagner; 1989, 1990;
Zelick and Narins, 1983) red
deer (Clutton-Brock et al.,
1979), and birds (Lambrechts
and Dhondt, 1988; Weary et al.,
1988,
1991). Most have concentrated
on the information contained in the temporal pattern of the signal. One would
also expect, however, that information would be contained in the size or power
of the signal and that this should interact with the temporal pattern. Thus,
there are two potential sources of variation that should be considered when
attempting to determine what information is transferred by these signals and
why they are performed repeatedly. The sequential assessment model (Enquist
and Leimar, 1983,
1987;
Enquist et al., 1990) suggests
that signals are repeated because receivers are able to make more accurate
assessments with each performance. The energetic (Payne and Pagel,
1996,
1997) and cumulative
(Payne, 1998) assessment
models, however, suggest that agonistic signals are repeated because the total
number of performances advertises stamina.
Hermit crabs interact in pairs in apparently agonistic encounters, at the
end of which there may be an exchange of shells. During the encounter, the
initiating crab hits its shell, in a series of bouts, against that of the
noninitiator ("shell rapping"). These bouts of rapping are
interspersed with pauses, during which the initiator pulls at the chelipeds of
the noninitiator and the latter may then evacuate its shell (see
Dowds and Elwood, 1983, for a
full description of shell fighting). These encounters are unusual in involving
two resources, and there is the possibility that both crabs could gain shells
more suited for their size. For example, a large crab in a shell that is too
small may exchange with a small crab in a shell that is too large
(Hazlett, 1978). The
possibility of mutual gain has given rise to the hypothesis that crabs
negotiate over the ownership of shells (Hazlett,
1978,
1983,
1987,
1989,
1996). This possibility
requires that the noninitiating crab assesses the quality of the initiating
crab's shell. Hazlett (1987)
suggested that shell rapping conveys this information in the fundamental
frequency of the individual raps. Recent work, however, has suggested that the
temporal pattern of shell rapping contains information about the motivation of
the initiator for shell acquisition and information concerning the relative
resource holding potential (RHP) of initiators
(Briffa et al., 1998). Because
aspects of the vigor of the pattern of rapping influenced whether or not an
eviction occurred, it was suggested that shell rapping is a repeated agonistic
signal of stamina (Briffa et al.,
1998) and that it is best described by the aggression model of
shell exchange (Dowds and Elwood,
1983,
1985;
Elwood and Glass, 1981;
Elwood et al., 1998).
The initial analysis of the temporal pattern of shell rapping
(Briffa et al., 1998) appears
to fit the energetic assessment model more closely than the sequential
assessment model. The impact of rapping, however, might vary with relative
RHP, and the likelihood of effecting an exchange should increase with the
impact of rapping. Furthermore, under the sequential assessment model, one
would expect that crabs that rapped with a high level of force should be able
to effect an exchange more quickly than those that rap with less force because
the stronger signal would indicate higher fighting ability. Under the
energetic assessment model, however, the force with which rapping is performed
should be less important because the level of advertisement is given by the
number of performances. The aim of this study is to examine the role of the
size of individual performances, in this case the energy of impact that the
raps achieve, and how this interacts with the timing of the signal.
In the present study the impact of rapping was reduced by placing a rubber coating on the part of the shell involved in shell rapping, whereas control initiators had another area of the shell coated with rubber. We predicted that if information concerning RHP or motivation is contained in the force of rapping, initiators supplied with rubberized shells should be less likely to effect an exchange than those supplied with control shells. Furthermore, we expected that when the impact of the raps is reduced, crabs should compensate by attempting to supply more force to each rap and/or by rapping more often. If a trade-off between the energy used in each rap and temporal vigor of signaling exists, either of these two strategies to overcome the effects of damping should affect the temporal pattern. Furthermore, crabs with a greater potential gain in shell quality may attempt to compensate for the damping more than would crabs with less potential gain.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Crabs were collected from the lower shore at Ballywalter, County Down, Northern Ireland, between April and July 1997. They were then held in groups of 75-100 in 60-cm x 30-cm plastic tanks, which were filled with aerated sea water to a depth of 10 cm, and fed twice weekly on chopped whitebait. Crabs were used within 1 week and then returned to the sea. We removed the crabs from their shells by cracking the shells open in a bench vice. We used only males in the experiments; females were supplied with new shells, and returned to the sea. This avoided sex differences in behavior that arise during the breeding season (Neil and Elwood, 1985
), which encompasses the time of the experiments. Only
crabs that were free from obvious parasites, loss of appendages, and recent
molt were used.
We allocated male crabs to pairs composed of a small crab and a large crab.
The smaller crab of each pair was supplied with a shell of the optimal weight
as determined by previous choice experiments
(Jackson, 1988) (100%
adequate) for the larger crab of that pair. In group 1 the larger crab of each
pair was supplied with a Littorina obtusata shell that was half the
optimal weight (50% adequate) and rubberized. In group 2 the larger crabs were
supplied with 50% adequate control shells. In group 3 the large crabs were
supplied with 25% adequate rubberized shells, and in group 4 with 25% adequate
control shells, thus providing for a higher motivation for the larger crabs to
exchange shells. Each crab was isolated in a 95-mm diameter crystallizing dish
for approximately 18 h before observations were made. Observations were
carried out in arenas consisting of a 95-mm diameter crystallizing dish
containing a 1-cm deep layer of washed sand and filled with 4 cm of aerated
sea water, inside an observation chamber fronted by a one-way mirror, such
that the crabs could not see the observer. The sand layer provided an adequate
grip for the crabs to walk and also prevented the shells from coming into
contact with the base of the area during rapping. Observations were started
when the small crab was placed into an arena containing the large crab of the
pair.
A two way ANOVA confirmed that there was no difference in relative weight
difference (RWD), calculated by
 |
Damping the raps was achieved by coating the external surface of the larger
crab's shells with an elastic material (Copydex, a latex-based adhesive). The
rubberized group had the rubber solution applied to the part of the shell that
makes contact with the opponent's shell during rapping; in the control group
the rubber was placed over the top of the shell such that it did not cover the
part of the shell that made contact. With normal shells, the energy of rapping
would be dissipated as sound and kinetic energy as the shell of the initiator
is brought into contact with the shell of the noninitiator. With rubberized
shells, however, some of the energy would be dissipated in compressing the
elastic coating of the initiator's shell, thus reducing the force of the raps.
To determine whether this method was effective, the sound intensity at the
point of impact was compared for crabs in rubberized and control groups. Sound
intensity is related to power (energy transferred/time) by
 |
 |
Data for the total number of bouts, total number of raps, mean number of
raps per bout, and the mean duration of pauses between bouts were not normally
distributed and were log10 transformed. We treated a gap of 
1 s
between two consecutive raps as a pause between two bouts of rapping.
Three-factor ANCOVAs were performed, the three factors being the treatment of
the initiator's shell, the outcome of the encounter, the percentage of
preferred shell weight of the shell supplied to the initiator, and the
regressor was relative weight difference. The degrees of freedom vary between
the different measures because different numbers of nonsignificant interaction
effects were removed from the different measures during calculation of the
ANCOVAs and because there were fewer replicates for the mean duration of
pauses than for the other measures because not all crabs performed more than
one bout of rapping.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Number of fights and exchanges
Out of 208 observations there were 107 fights, of which 83 resulted in exchanges, whereas the initiator gave up in the remaining 24 fights. There was no difference according to treatment or shell size in the likelihood of a fight being initiated. There was a nonsignificant trend for crabs supplied with 50% adequate shells that were rubberized to be less likely to effect an exchange than those supplied with 50% adequate control shells and a similar nonsignificant trend for crabs supplied with 25% adequate shells. When both groups are pooled, however, initiators supplied with control shells were significantly more likely to effect an exchange (
12
= 4.7, p <.05; Table
1). There was a further a nonsignificant trend for those in 25%
adequate shells to be more likely than those in 50% adequate shells to effect
an exchange.
|
The change in shell adequacy following an exchange for the noninitiators
covaries with the size difference between the two crabs, and this is an
unavoidable consequence of these interactions. To determine the change in
shell quality for non-initiators, the percent change in deviation from
preferred shell size (%CD) (see Briffa et
al., 1998) after an exchange was calculated. The %CD values ranged
from 41.6% to 103.5% (mean ± SE = 72.2% ± 1.425). Thus shell
quality for the noninitiators was always decreased by an exchange whereas
initiators always gained a shell that was closer to their preferred size.
Mean number of raps per bout
Crabs supplied with rubberized shells performed fewer raps per bout than
those supplied with control shells (F1,98 = 14.61,
p <.0002). The mean number of raps per bout, however, did not
differ according to the outcome of the encounter or to the size of shell
supplied to the initiator, and RWD had no effect on the mean number of raps
per bout. There were no interaction effects between treatment and outcome or
between outcome and shell size. There was, however, a significant interaction
between treatment and shell size, with crabs in rubberized shells showing
little difference with shell size, whereas for crabs in control shells there
was a marked effect due to shell size, with those in 25% adequate shells
giving the greater number of raps per bout (F1,98 = 4.77
p <.05; Figure 1).
There was no significant three-factor interaction.
|
To determine how the number of raps performed per bout varied as the fight
progressed, a three-factor ANCOVA, including one repeated measure, was
performed. The repeated measure was the number of raps performed in each of
the first six bouts by crabs that performed at least seven bouts of rapping
(bout number). This reduced the sample size to 63. The seventh bout was
omitted from the analysis to avoid the unusually high or low number of raps
which can occur in the last bout of rapping, depending on the outcome of the
encounter (Briffa et al.,
1998). The two other factors were shell size and treatment, and
the regressor was RWD.
The number of raps per bout increased with the RWD between the two crabs (F1,58 = 4.14, p <.05), and crabs supplied with 25% adequate shells performed more raps per bout during the first six bouts of the fight than did those supplied with 50% adequate shells (F1,58 = 10.1, p <.002). Crabs supplied with rubberized shells performed fewer raps per bout (F1,58 = 5.2, p <.05), and there was a significant interaction effect between bout number and treatment (F5,290 = 2.5, p <.05; Figure 2). Crabs supplied with rubberized shells performed a consistently low number of raps in each successive bout, whereas those supplied with control shells performed fewer raps in each consecutive bout.
|
Mean duration of pauses
The mean duration of pauses between bouts of rapping was not affected by
the treatment of the shell or by the size of the shell supplied to the
initiator. Initiators that effected an exchange, however, left shorter pauses
between bouts than those that did not effect an exchange
(F1,98 = 9.32, p <.005;
Figure 3). RWD had no effect on
the mean duration of pauses between bouts of rapping, and there were no
interaction effects.
|
Duration of encounters
Crabs supplied with rubberized shells performed more bouts of rapping than
those in the control group (F1,95 = 6.00, p
<.05) and initiators supplied with 25% adequate shells performed more bouts
than those supplied with 50% adequate shells (F1,95 =
3.51, p <.05). There was no difference in the total number of
bouts performed between the two outcomes. Relatively large initiators
performed more bouts of rapping than relatively small initiators
(F1,95 = 4.58, p <.05). There was a
significant interaction between treatment and outcome, with initiators
supplied with rubberized shells that did not effect an exchange performing
more bouts than those that did exchange, whereas initiators supplied with
control shells that did not effect an exchange performed fewer bouts than
those that did exchange (F1,95 = 7.80, p
<.01). There were no significant interactions between outcome and shell
size or between treatment and shell size. There was no three-factor
interaction.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The key finding is that crabs in rubberized shells are less likely to evict the noninitiator from its shell. Furthermore, the temporal pattern of rapping is disrupted when the force of rapping is reduced. The rubberized shells clearly prevented the initiator from achieving the normal impact when shell rapping, and thus we conclude that the impact is an important variable in shell fights. This is congruent with the finding that naked crabs initiate fights and rap their soft abdomens against the shells of noninitiators, with little impact, and are unlikely to effect an exchange (Elwood and Glass, 1981
). In
normal fights, impact of rapping would vary with the effort with which it is
performed and thus might contain information about the initiator. This
information may be about RHP and/or motivation and could thus signal likely
future costs to the noninitiator should it persist in the fight. Another
possibility, however, is that rapping might also have direct effects on the
ability of noninitiators to maintain an adequate grip on their shells. The
magnitude of these effects would normally be expected to vary with the effort
with which the raps are performed.
It is difficult, using the present data, to determine whether these
interactions are best described by the sequential assessment model (Enquist
and Leimar, 1983,
1987,
Enquist et al., 1990) or by
alternative models. Previous analysis of the temporal pattern of shell rapping
(Briffa et al., 1998) suggested
that RHP was signaled by the cumulative effect of all of the repeated
performances, in line with the energetic model proposed by Payne and Pagel
(1997). The present finding
that the duration of encounters is increased when initiators are supplied with
rubberized shells is congruent with the sequential assessment model because
the reduced force of raps would indicate poor fighting ability and thus
noninitiators would be less likely to give up early in the fight. It is also
conceivable, however, that this effect would be expected under the energetic
model (Payne and Pagel, 1997)
if information about RHP is contained in the force of rapping as well as in
the temporal pattern.
Elwood and Neil (1992)
suggested that shell rapping could reduce the ability of noninitiators to
maintain an adequate grip on their shells by causing disorientation or by
disrupting the respiratory flow of water over the gills, causing oxygen
impoverishment. Presumably, the effects of both of these possibilities would
increase with the impact of rapping. Work by Chapple
(1993),
1997) suggests that vibrations,
in excess of an amplitude threshold but within a specific range of
frequencies, affect the reflex contraction of abdominal muscles and reduce the
crab's grip in the shell. Thus crabs in rubberized shells would also be
expected to be less effective in fights under this model of direct
effects.
Because the energy used and hence the impact of rapping appears to be an important feature of this activity, one would expect that initiators that are prevented from rapping with their usual level of impact would attempt to compensate in some way. They could either produce more raps per bout or increase their effort with each rap in an attempt to hit harder to overcome the effects of the rubber. The data, however, do not support the former possibility. Initiators supplied with rubberized shells performed fewer raps per bout than those supplied with control shells. Perhaps then the crabs are attempting to hit harder in a vain attempt to compensate. If this is the case, then each rap would now take more energy to deliver and thus the crabs would be subjected to higher levels of fatigue and be unable to use as many raps per bout. The idea that fewer raps per bout indicates higher levels of fatigue is supported by the finding that crabs supplied with 25% adequate shells performed more raps per bout than those occupying the heavier 50% adequate shells because the lighter shells may be rapped with the use of less energy. Furthermore, there is a marked decline in the number of raps per bout as fights progress (Figure 2), again suggesting that the number of raps per bout is regulated by fatigue. Thus, the marked change in the temporal pattern of rapping may be caused by the heightened levels of fatigue because of an increase in the effort that they put into their raps.
The fact that the difference in number of raps per bout between initiators
supplied with rubberized and control shells is apparent even in the first bout
(Figure 2) suggests that
initiators are able to assess the effects of their raps during bouts, rather
than having to wait for pauses in order to gather this information. Perhaps
they are able to assess the impact of their raps by using the reverberations
within their own shells. This leads to a further possibility that the
difference in the vigor of rapping between the two groups of crabs is due to
differing motivation. Those crabs supplied with rubberized shells could become
less motivated as the fight progresses because of the lower expected pay-off
which results from the decreased effects of their raps. Although theory
predicts that they should not advertise this decrease in motivation, analyses
of fights in cichlid fish (Turner and
Huntingford, 1986) and in hermit crabs
(Briffa et al., 1998) suggests
that the participants in agonistic interactions may reveal their likely future
actions.
We have shown previously that shell rapping is a costly activity for the
crab performing the action (Briffa et al.,
1998) and that different aspects of the temporal vigor are traded
off against each other. Thus, pauses between bouts are traded off against the
number of raps in the following bout (Briffa and Elwood, in press).
Furthermore, crabs leaving short pauses between bouts
(Figure 3) and performing
numerous raps per bout (Briffa et al.,
1998) are more likely to effect an eviction. Shell rapping,
therefore, is a complex activity, and although we conclude that impact has an
effect on the lower probability of eviction, we cannot rule out the
possibility that eviction is also influenced by the reduced number of raps per
bout. To tease out the influence of each aspect of rapping (power of each rap,
number of raps per bout, repetition rate within a bout, and duration of pauses
between bouts) and how these features change during the course of a fight and
interact, all must be measured for each encounter. To date, however, they have
only been measured in separate studies (Briffa and Elwood, in press;
Briffa et al., 1998).
Under the negotiation model of shell exchange (Hazlett,
1978,
1983,
1987,
1989,
1996), shell rapping provides
noninitiators with information about the shell of the initiator. It has been
suggested that noninitiators assess the fundamental frequency of rapping,
which is related to the volume of the initiator's shell. The fundamental
frequency of rapping should not vary with impact, and the negotiation model
does not predict an effect of impact on the likelihood of an eviction, except,
perhaps, if the impact is so small as to be undetectable by the noninitiator.
Because crabs supplied with rubberized shells can effect an exchange, it is
unlikely that this was the case in the fights reported here. The idea that
crabs which hit harder are more likely to effect an exchange (either because
the impact contains information or by causing detrimental effects directly),
however, is congruent with the aggression model of shell exchange (Dowds and
Elwood, 1983,
1985;
Elwood and Glass, 1981;
Elwood et al., 1998), which
views shell fighting as a war of attrition with the eventual victor being the
participant prepared to commit the most time and energy to the contest. Note
that the cumulative assessment model
(Payne, 1998) predicts similar
results to the war of attrition model. Furthermore, one would expect
cooperative signals, such as that suggested by the negotiation model, to
utilize low energy systems. Krebs and Davies
(1991) pointed out that
receivers should evolve high sensitivity to such signals and that natural
selection would then favor senders that saved energy by expending minimal
amounts on performing the signal. Shell rapping, however, appears to be
dependent on raps of high energy.
These data suggest that the normal force with which shell rapping is
performed is an important feature of the behavior and, in conjunction with the
temporal vigor of rapping, affects the likelihood of an exchange. It is not
clear, however, whether this is because the force of rapping contributes to
the information contained in the timing of the signal or because the raps
directly affect the ability of the noninitiator to maintain an adequate grip
of its shell. It is possible that shell rapping can act both as a signal and
cause direct detrimental effects for the receiver. In this sense the activity
would be a tactical behavior (Bradbury and
Vehrencamp, 1998), which affects the condition of the noninitiator
and, therefore, changes the appropriateness of the two courses of action open
to it (i.e., give up or persist with defense). Enquist
(1985) and Grafen
(1990) have suggested that a
given activity could have both signal and tactical functions. Thus shell
rapping could simultaneously transfer information concerning initiator
motivation and relative RHP and cause direct detrimental effects.
|
ACKNOWLEDGEMENTS |
|---|
We thank the Department of Education for Northern Ireland and the BBSRC for funding this work. We are grateful to two anonymous referees and Magnus Enquist, whose comments helped us to make considerable improvements to the manuscript.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Bradbury JW, Vehrencamp SL, 1998. Principles of animal communication.. Sunderland, Massachusetts: Sinauer Associates.
Briffa M, Elwood RW, Dick JTA, 1998. Analysis of
repeated signals during shell fights in the hermit crab Pagurus
bernhardus. Proc R Soc B 265:
1467-1474.
Briffa M, Elwood RW, 2000. Analysis of the fine scale timing of repeated signals: does shell rapping in hermit crabs signal stamina? Anim Behav 59: 159-185.[Web of Science][Medline]
Chapple WD, 1993. Dynamics of reflex cocontraction in
hermit crab abdomen—experiments and a systems model. J
Neurophysiol 69:
1904-1917.
Chapple WD, 1997. Regulation of muscle stiffness
during periodic length changes in the isolated abdomen of the hermit crab.
J Neurophysiol 78:
1491-1503.
Clutton-Brock TH, Albon SD, Gibson RM, Guinness FE, 1979. The logical stag: adaptive aspects of fighting in red deer (Cervus elephus L.). Anim Behav 27: 211-225.
Dowds BM, Elwood RW, 1983. Shell wars: assessment strategies and the timing of decisions in hermit crab shell fights. Behaviour 73: 64-76.
Dowds BM, Elwood RW, 1985. Shell wars II: the influence of relative size on decisions made during shell fights. Anim Behav 33: 649-656.
Elwood RW, Glass CW, 1981. Negotiation or aggression during shell fights of the hermit crab Pagurus bernhardus. Anim Behav 29: 1239-1244.
Elwood RW, Neil SJ, 1992. Assessments and decisions: a study of information gathering by hermit crabs. London: Chapman and Hall.
Elwood RW, Wood KE, Gallagher MB, Dick JTA, 1998. Probing motivational state during agonistic encounters in animals. Nature 393: 66-68.
Enquist M, 1985. Communication during aggressive interactions with particular reference to variation in choice of behaviour. Anim Behav 33: 1152-1161.[Web of Science]
Enquist M, Leimar, O, 1983. Evolution of fighting behaviour: decision rules and assessment of relative strength. J Theor Biol 102: 387-410.
Enquist M, Leimar O, 1987. Evolution of fighting behaviour: the effect of variation in resource value. J Theor Biol 127: 187-205.
Enquist M, Leimar O, Ljungberg T, Mallner Y, Segerdahl N, 1990. A test of the sequential assessment game: fighting in the cichlid fish Nannacara anomala. Anim Behav 40: 1-14.
Grafen A, 1990. Biological signals as handicaps. J Theor Biol 144: 517-546.[Web of Science][Medline]
Hazlett BA, 1978. Shell exchange in hermit crabs: aggression, negotiation or both? Anim Behav 26: 1278-1279.
Hazlett BA, 1983. Interspecific negotiations: mutual gain in exchanges of a limiting resource. Anim Behav 31: 160-163.
Hazlett BA, 1987. Information transfer during shell exchange in the hermit crab Calcinus tibicen. Anim Behav 35: 218-226.
Hazlett BA, 1989. Shell exchanges in the hermit crab Calcinus tibicen. Anim Behav 37: 104-111.
Hazlett BA, 1996. Assessments during shell exchanges by the hermit crab Clibanarius vittatus, the complete negotiator. Anim Behav 51: 567-573.
Jackson NW, 1988. Information gathering and decision making during shell selection by the hermit crab Pagurus bernhardus (PhD dissertation). Belfast. The Queen's University of Belfast.
Krebs JR, Davies NB, 1991. Behavioural ecology: an evolutionary approach, 3rd ed. Oxford: Blackwell.
Lambrechts MM, Dhondt AA, 1988. The anti-exhaustion hypothesis: a new hypothesis to explain song performance and song switching in the great tit. Anim Behav 36: 327-334.
Lopez PT, Nairns PM, Lewis ER, Moore SW, 1988. Acoustically induced call modification in the white-lipped frog, Leptodactylus albilabris. Anim Behav 36: 1295-1308.
Neil SJ, Elwood RW, 1985. Behavioural modifications during egg brooding in the hermit crab Pagurus bernhardus L. J Exp Biol Ecol 94: 99-114.
Payne RJH, 1998. Gradually escalating fights and displays: the cumulative assessment model. Anim Behav 56: 651-662.[Web of Science][Medline]
Payne RJH, Pagel M, 1996. Escalation and time costs in displays of endurance. J Theor Biol 183: 185-193.
Payne RJH, Pagel M, 1997. Why do animals repeat displays? Anim Behav 54: 109-119.[Web of Science][Medline]
Turner FG, Huntingford FA, 1986. A problem for game theory analysis: assessment and intention in male mouthbrooder contests. Anim Behav 34: 961-970.
Wagner WEJ, 1989. Graded aggressive signals in Blanchard's cricket frog: vocal responses to opponent proximity and size. Anim Behav 38: 1025-1038.
Wagner WEJ, 1992. Deceptive or honest signalling of fighting ability? A test of alternative hypotheses for the function of changes in call dominant frequency by male cricket frogs. Anim Behav 44: 49-462.
Weary DM, Krebs JR, Eddyshaw R, McGregor PK, Horn A, 1988. Decline in song output by great tits: exhaustion or motivation? Anim Behav 36: 1242-1244.
Weary DM, Lambrechts MM, Krebs JR, 1991. Does singing exhaust male great tits? Anim Behav 41: 540-542.
Zelick RD, Nairns PM, 1983. Intensity discrimination and the precision of call timing in two species of neotropical tree frogs. J Comp Physiol 30: 728-733.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M. Briffa Signal residuals during shell fighting in hermit crabs: can costly signals be used deceptively? Behav. Ecol., May 1, 2006; 17(3): 510 - 514. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Briffa, R. W. Elwood, and J. M. Russ Analysis of multiple aspects of a repeated signal: power and rate of rapping during shell fights in hermit crabs Behav. Ecol., January 1, 2003; 14(1): 74 - 79. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||