BEHAVIOR AND CNS INTEGRATION 219 



able to learn and retain classically conditioned reflexes to light onset 

 (Karamian 1956). The misdirected approach behavior seen in subjects with 

 substantial damage to the central telencephalic nuclei argues for the presence 

 of an orientation deficit. Moreover, it appears to be specific to localizing 

 visual stimuli since preliminary results indicate that such sharks can still 

 localize low-frequency, pulsed sound sources (Thomas and Jane, personal 

 communication). 



While our current knowledge does not permit a final conclusion about the 

 nature of the observed behavioral deficit, there is no doubt that the behav- 

 ioral evidence supports the corresponding anatomical and electrophysiolo- 

 gical data in affirming the presence of telencephalic mechanisms in the shark 

 central visual system. 



An Integrated View of Central Visual Components 



The classic belief in tectally dominated vision in sharks at least had the 

 virtue of simplicity. We must now try to understand the functional interre- 

 lationships that govern the operation of several central visual system com- 

 ponents. While this chapter has dwelt on two of the most prominent, the 

 tectum and the central telencephalic nuclei, other visual regions need to be 

 addressed. Most of these are in the thalamus, which has been shown (Ebbes- 

 son and Ramsey 1968, Graeber and Ebbesson 19726) to have three distinct 

 areas receiving retinal terminations: the dorsomedial optic nucleus or pre- 

 tectal area, the dorsolateral optic complex, or lateral geniculate nucleus 

 (Houser 1901), and the ventrolateral optic nucleus. These three areas are also 

 connected anatomically with the tectum and telencephalon (Ebbesson 

 1972a, 19726, Schroeder and Ebbesson 1974), but at the present time it is 

 not known for certain which of the three contribute to the visual input 

 reaching the latter. The presence of a posterior accessory optic tract, similar 

 to that seen in mammals, has also been reported in the thalamus, but only in 

 tiger sharks (Ebbesson and Ramsey 1968). Finally, in all species examined so 

 far, there are a small number of retinal fibers that leave the optic tract just 

 after it decussates and enter the contralateral hypothalamus, where they may 

 be involved in the control of endocrine function and circadian rhythms. 



Some work that begins to separate the functional interrelationships 

 between these different central visual areas has already been described. 

 Specifically, Veselkin and Kovacevic (1973) found that telencephalically 

 evoked responses to optic nerve stimulation were retained in rays after com- 

 plete destruction of the midbrain optic tectum. It appears therefore that the 

 retinal input to the dorsolateral thalamus, rather than the overlapping tectal 

 input, is being relayed to the contralateral central telencephalic nucleus. This 

 finding may explain why we found no serious impairment in visual discri- 

 minative learning from tectal lesions. Both the physiological and behavioral 

 results suggest that there may be two functionally different subsystems in 

 the overall elasmobranch central visual system: a midbrain retinotectal sys- 

 tem and a forebrain retino-thalamo-telencephalic system. Furthermore, these 

 subsystems may be partly redundant, as indicated by the extent of 



