also indicate an increased capability for multi-modal sensori-motor 

 integration. Hence, it is not surprising to the neurobiologist that 

 shark behavior is apparently less stereotyped than often conceived on tht 

 basis of the shark's traditionally primitive simple brain. The naivete 

 of this old notion is further pointed out by recent field studies 

 (Johnson & Nelson, 1973) and Baldridge's (1974) analyses which have 

 demonstrated the variability and unpredictability of shark attacks on 

 humans. 



Similar studies on the central neural mechanisms integrating vibratory 

 tactile, electrical, and olfactory cues into response patterns are sorely 

 needed and now appear feasible. Some such work has been initiated by 

 Roberts, who has investigated the motor end of the stimulus response 

 chain in dogfish. (Roberts, 1969a, 1969b; Roberts & Russell, 1972). 



Another intriguing and promising approach to brain-behavior studies 

 in sharks has been suggested by Okado and his colleagues (1969). Their 

 delineation of the substantial morphological differences among the brains 

 of different species strongly argues for a comparative strategy to answer 

 some of the questions about the relationships between brain structure and 

 behavior. This has yet to be tried amidst a laboratory setting. 



As might be expected from the newly discovered complexity of the shark 

 forebrain, shark behavior has recently been shown to be more flexible than 

 previously thought. A number of learning experiments have demonstrated 

 that both lemon and nurse sharks can be trained as easily as many mammals 

 and can retain learned tasks for a considerable period of time (Aronson, 

 1963; Aronson et al, 1967; Clark, 1959; Graeber, 1972; Graeber & Ebbesson, 

 1972). The importance of such learning studies should not go unrecognized. 

 Much of the progress which has been made in relating brain structure and 

 physiology to behavior in mammals has resulted from experiments employing 

 learned tasks to assess the effects of experimentally varying central 

 neural states. It is now possible to apply similar behavioral technology 

 to learning more about shark brain function. Moreover, the learning data 

 underscore the important contribution of past experience in determining 

 the momentary response tendencies of these animals. Consequently, it can 

 be expected that experiential factors will influence the effectiveness of 

 any technologies intended to thwart or predict shark behavior, especially 

 in a given locale. 



The hypothalamus and related areas of the brain have been shown to be 

 the focal point for arousal or motivational systems related to feeding 

 and attack in fishes as well as other vertebrates. For this reason a 

 separate section dealing with this important topic in sharks and bony 

 fishes is included below. 



2) Neural Mechanisms of Feeding and Aggression 



Feeding patterns have been described for many sharks maintained in 

 captivity and in the wild (Clark, 1963; Gilbert, 1970; Graeber, 1974; 

 Hodson et al . , 1967; Kalmijn, 1971; Randall, 1963; Springer, 1967; Tester, 

 1963) . The presence of chemical stimuli such as are released from freshly 

 killed animals can cause considerable attraction of sharks and may result 

 in a so-called "feeding frenzy" in which sharks are known to bite at any- 

 thing that moves. Some of the sharks themselves may be eaten during this 



22 



