The Agonistic Behaviour of the Velvet Swimming Crab, Liocarcinus puber (L.) (Brachyura, Portunidae)

The velvet swimming crab, Liocarcinus puber (L.), is common in shallow, rocky sublittoral areas on the Atlantic coasts of Europe. In Britain, the commercial importance of L. puber has increased in the last 10 years, following the collapse of the Spanish fishery for this species. L. puber has traditionally been regarded as an "aggressive" crab, mainly due to its rapid defensive reaction to humans, but little is known of its intraspecific agonistic behaviour. The aims of the work in this thesis were: to describe the intraspecific agonistic behaviour of L. puber in relation to the size of interacting crabs; to investigate the initiation, outcome and content of agonistic interactions in competition for different resources; to estimate the energetic cost of agonistic interactions; to gauge the importance of this behaviour in natural populations and to determine whether behaviour observed in the laboratory is representative of that in the field; and to assess the influence of agonistic behaviour on the capture efficiency of creels. The agonistic behaviour of L. puber has been compared with predictions from game theory. The agonistic displays of L. puber were similar to those described for other portunids, particularly the congeneric, L. depuralor. The relative size of interacting crabs had a major influence on the content, duration and outcome of interactions. However, the smaller crab was as likely to initiate interactions as the larger crab, except when there was a large size difference, when all interactions were initiated and won by the iarger crab. As the size difference decreased, the proportion of interactions won by the smaller crab approached 50% and interactions became longer, involving more potentially injurious behaviour. There was little apparent effect of absolute size on interactions, other than a slight reduction in interaction duration with increasing crab size. The effects of resource value on agonistic behaviour have been investigated in the context of competition for food and mates. Over 5 days of food deprivation, interactions became more intense, involving a greater proportion of potentially injurious behaviour. After 12 days however, there was either only a slight further increase, or a reduction in intensity when the odour of food was absent or present, respectively. The increasing costliness of interactions during the initial period of deprivation accords with the predictions of game theory. Possible reasons for the relative reduction in intensity after further food deprivation are discussed. Males compete vigorously for sexually receptive females in the laboratory and there was indirect evidence that this occurs in the natural situation. Laboratory observations indicated that competition between males for females involves a high risk of injury and that males could successfully defend females against larger males. This result was reflected in studies of agonistic interactions between males exposed to the odour of sexually receptive females. In that situation, smaller males were as likely to win as their larger opponents, in contrast to the advantage that larger crabs usually have. Exposing males to the odour of receptive females prolonged interactions, but exposing them to the odour of conspecifics of either sex resulted in more potentially injurious behaviour. The energetic cost of agonistic interactions has been investigated by using the scaphognathite rate as an indicator of the oxygen consumption of interacting crabs. There was no significant anaerobic metabolism during agonistic behaviour. Extremely high and variable respiratory rates were recorded during agonistic interactions. The maximum scaphognathite rates recorded were related to the degree of escalation. Estimation of the energetic cost of interactions took account of the protracted recovery period. The magnitude of this estimate was related to the content and duration of the interaction for losers, but not winners, although there was no significant difference between losers and winners in this measure. This study also highlighted differences in behaviour between winners and losers following interactions, in the laboratory at least. Field observations by diving indicated that L. puber are inactive for much of the time. Males were more abundant and more active than females, suggesting that the majority of intraspecific encounters in the field are between males