M. Briffa et al., ANALYSIS OF REPEATED SIGNALS DURING SHELL FIGHTS IN THE HERMIT-CRAB PAGURUS-BERNHARDUS, Proceedings - Royal Society. Biological Sciences, 265(1404), 1998, pp. 1467-1474
Shell exchanges between hermit crabs may occur after a period of shell
rapping, when the initiating or attacking crab brings its shell rapid
ly and repeatedly into contact with the shell of the non-initiator or
defender, in a series of bouts. There are two opposing models of hermi
t crab shell exchange and the function of shell rapping. The negotiati
on model views shell exchange as a mutualistic activity, in which the
initiator supplies information about the quality of its shell via the
fundamental frequency of the rapping sound. The aggression model views
shell rapping as either detrimental to the defending crab, or as prov
iding it with information about the initiator's ability or motivation
to continue, or both. The negotiation model makes no predictions about
the temporal pattern of rapping, but under the aggression model it wo
uld be expected that crabs that rapped more vigorously would be more l
ikely to effect an exchange. Repeating the signal could be expected un
der either model. Crabs that achieve an exchange rap more vigorously,
rapping is more persistent when a clear gain in shell quality may be a
chieved, and the vigour is greater when the relative resource-holding
potential (or 'fighting ability') is high. These findings support the
aggression model rather than the negotiation model. Contrary to the pr
edictions of game theory, crabs that do not effect an exchange appear
to signal that they are about to give up. The data suggest that rappin
g is performed repeatedly because the accumulation of all of the perfo
rmances acts as a signal of stamina.