Behavioural and physiological studies of fighting in male Tilapia Zillii (Cichlidae)
This thesis is a compilation of several behavioural and physiological studies of fighting in a species of African cichlid fish, Tilapia zillii. The behavioural studies are concerned with explaining the functional significance and behavioural organisation of fighting. The physiological studies are concerned with attempting to elucidate the mechanisms that underlie the expression of aggressive behaviour. My aim is show how these two levels of study are related and why it is important to maintain a balance between them in the study of animal behaviour. Chapter 2 reports a senes of laboratory experiments that investigated how asymmetries in body size and gonadal state influence the fighting strategies of male T. zillii, Fights between animals over limited resources often end in victory to the larger contestant. Game theory predicts that relative body size is assessed during the fight and thus also determines fight duration and intensity. However, if the contestants differ in the value they place on the disputed resource, this can override the effects of relative body size. In the experiment described here, relative gonad weight was a much stronger predictor of fight outcome than relative body size. This suggests that males with large gonads fight harder to defend their territory, implying that the value of a territory is a function of the gonadal state of the individual. Despite this, a detailed behavioural analysis suggested that relative body size is assessed through a behaviour termed 'mouth wrestling'. In addition to persisting in the fight, smaller contestants escalated to a higher degree (in terms of biting), especially if they went on to win the tight. Winners and losers differed consistently with regard to a behaviour termed mouth locking, suggesting that through this behaviour the fish were assessing an asymmetry related to the intention to persist with the fight, although I could not demonstrate that this was related to relative gonad weight. Chapter 3 was undertaken with the aim of gaining information on the social structure and occurrence of aggressive behaviours in a semi-natural environment. To this end, a study was made of the social interactions of a group of male fish allowed free range over approximately ten metres of artificial 'river bed' during a four week period. Under these conditions, the social structure of T. zillii may be described as a loose group with an unstable dominance hierarchy. Aggression is a major component of this species' time budget, with different acts and degrees of escalation probably functioning to establish, challenge and maintain dominance relationships within the hierarchy. The expression of territorial behaviour appeared to be conditional upon social status (only the two top ranked males were seen to dig nests and consistently court females). One escalated fight was observed between the two highest ranked fish, but it was not clear whether this was a territorial dispute, a challenge for the top position, or a dispute over both. Thus, while the immediate reasons for escalated fighting between male T. zillii depends upon the social and physical environment, it is probably ultimately caused by competition for mating opportunities. Chapter 4 deals with the issue of how body size and gonadal state relate to life-history aspects of the animal by means of a morphometric analysis. The analysis provided evidence to suggest that there was a trade off between gonad size and the storage of fat It is not clear whether this relationship arises from a direct energetic trade-off or is a result of adopting different behavioural strategies. I also explored the relationship between behaviour, morphology and somatic and reproductive condition. It was possible to predict gonadal state on the basis of mouth morphology which may explain why so much fighting involves use of the mouth in this species. In Chapter 5 I studied the proximate costs of fighting in terms of physical injury and the metabolic consequences of engaging in a energetically demanding activity. In relation to injuries incurred during fights, losers suffered greater scale loss than winners, especially if the loser was larger than its opponent. In relation to energy metabolism, fighting resulted in significant depletion of total sugar reserves from the muscle and the liver (compared to unfought controls). It appears that the muscle energy reserves are respired anaerobically, as was evident from the accumulation of lactate in the muscle. Interestingly, losers had significantly higher levels of lactate than winners. Together, the data on injury and metabolic state suggest that fighting is costly for both winners and losers, but that this is especially marked for losers. These data are discussed in relation to models of animal decision-making and it is concluded that the summation of different proximate costs incurred during fighting is likely to underlie the making of decisions such as continuing, giving up or escalating the fight. In Chapter 6 I report a study of plasma concentrations of gonadal steroids in T. zillii in relation to fighting and gonadal state. The gonadal steroids, particularly the androgens, have been shown to be associated with aggression in a wide range of species and seemed like obvious candidates for the mechanism by which gonadal state influences behaviour Blood samples were taken from fish immediately after fighting and the plasma concentrations of the following gonadal steroids were then determined by radioimmunoassay: testosterone (T), l l-ketotestosterone (1IkT) and 17a,20~- Dihydroxy-4-pregnen-3-one (17,20-P) and estradiol (Ez). T concentrations did not correlate with GSI and no differences were detected between winners, losers and unfought controls. II kT concentrations were on average twice those of T and negatively correlated to GSI, although winners, losers and controls were not significantly different. T and II kT concentrations were positively correlated with each other but 17,20-P and Ez levels were too low to be accurately measured. It appears that in T. zillii, II kT is the major androgen as it is in most other mature male teleosts. While the present results suggest that these sex steroids may play a role in physiological regulation of testicular maturation, they do not support the idea that these steroids are the mechanism by which GSI influences aggressive behaviour. In Chapter 7 I review the main conclusions of the previous chapters. I then offer my personal opinion on how the different levels and approaches taken throughout the study interrelate and collectively reinforce each other and why integration between disciplines is important in the study of animal behaviour.