The South African demersal longline experiment is assessed with emphasis on the target species, kingklip Genypterus capensis. The hypothesis that kingklip on the South African coast comprise a unit stock is tested. Recommendations for the management of the fishery are made. Longlining was found to be commercially viable. Techniques have been developed to target on either kingklip or the Cape hakes, Herluccius capensis and H. paradoxus. Longline fishermen exploit the kingklip spawner stock from August to No v ember by taking advantage of prespawning aggregations on the South-East Coast . Effort switches to the West Coast in late summer and early winter where kingklip are less abundant and a larger proportion of hake is caught. Catch rates of kingklip on the South Coast have declined sharply and the resource there has been exploited at a rate greater than that required to retain 50% of the unexploited biomass . There is a 17,6% probability that the spawner biomass on the South Coast has already been depleted below its pristine level. On the West Coast the catch rates of kingklip are lower than on the South Coast and have not changed significantly, although within the 95% confidence limits there is a possibility that the resource there has also been depleted below 50% of its pristine level. Genypterus capensis on the South African coast comprise a unit stock. Comparison of kingklip morphology using multivariate and discriminant function analysis and of otolith morphology using univariate statistics shows that there are no significant differences between the fish on the West and South Coasts. They are a slow growing species and were aged up to 25 years. Kingklip on the West Coast are smaller and have a lower L~ than those on the South Coast, but this is not a characteristic of discrete stocks. Kingklip on the South Coast mature earlier than those on the West Coast and aggregate to spawn on the South - East Coast in spring. There is no obvious spawning period on the West Coast. It is hypothesized that kingklip spawning on the South-East Coast is in response to favourable environmental conditions that enhances the survival of their eggs and larvae. It is recommended that demersal longlining be established as a permanent kingklip-directed fishery but that effort should not be allowed to increase. The West and South Coasts should be managed .separately. A TAC of 5 OOOt for kingklip should be introduced for 1989 of which 2 OOOt should be allocated to the West Coast and 3 OOOt to the South Coast . A closed season for kingklip from 1 August to 30 September on the South-East Coast is recommended. Hake-directed longlining should not be allowed as its effect on the hake spawner stocks are unknown and could be a potential threat to the stability of the demersal trawl fishery.

Ophidiidae are major benthopelagic fishes with wide distribution and depth range. Three types of sonic structural patterns have been described in some studied ophidiid fishes, however, the significance of the different types of the sonic apparatus, and the distribution in the subfamilies remain unclear. In addition, although the sonic apparatus and the sonic mechanisms of the high frequency sounds produced by cusk-eels has been explained, the characteristics of extrinsic swimbladder sonic muscle is still unknown. In this study, I investigated three ophidiid subfamilies from Taiwan and reviewed the published data to study the morphological traits of their sonic apparatus. Inaddition, Hoplobrotula armata were related to study the proteomic characterization of the extrinsic swimbladder sonic muscle. Results indicate that the 13 ophidiid species in this study could be grouped into five sonic structural patterns, and three subfamilies could be separated accordingly. Furthermore, because the sister group of the family Ophidiidae-carapidae is soniferous, I suggest that the common ancestor of the three ophidiid subfamilies should also be soniferous. Comparisons made in regarded to environmental factors indicate that diversity of sonic apparatus in three ophidiids subfamiliesis depth dependent. The extrinsic swimbladder sonic muscles could be separated into ventral sonic muscle and intermediate sonic muscle. The histological cross-sections of the ventral sonic muscle fibers show peripheral ring myofibrillar region and central core and they are similar with the sonic muscle. Ventral muscle weights were higher in males than females, but the muscle fibers are smaller in males. Protein values of ventral sonic muscles showed high expression in fast and long duration constructional proteins, and males were higher in protein expression than females. Intermediate sonic muscles, on the other hand, were larger in females than males. The cross-sections of muscle fibers were similar to the white muscle. Expression in the fast constructional related protein in the intermediate sonic muscle was higher than ventral sonic muscles, and the metabolically related protein was lower than ventral sonic muscles. The phylogeny of ophidiiform fishes is not clear at present; the sonic-apparatus diversity in ophidiid fishes observed in this study becomes useful to reveal the relationship phylogenetic of ophidiids fishes. Regard the physiology of sound production, I suggest that the larger ventral sonic muscle in males are be provides a better constructional ability, and their smaller fibers are adaptative for energy metabolism regarded for continuous fast constraction and fatigue resistance. The longer intermediate sonic muscle in females is a better constructional ability than male. In intermediate sonic muscle were be fast constructional related protein, higher in expression than the ventral sonic muscles, the metabolic related protein was lower than the ventral sonic muscles. These results suggest that the intermediate sonic muscle has a hight constructional ability but has a disventage of being easily fatigue.