Gibbons et a\ Habitat use by demersal nekton on the continental shelf in the Benguela ecosystem 



487 



North 



Sole 



Kingklip 



Goby 



False Jacopever 



Hake 



South 



-H 1 1 



10. 20. 30. 40. 50 60 70. 80. 90. 100. 

 Bray-Curlis similarity 



Cuttlefish 



Goby 



Grenadier 



Jacopever 



Sole 



Dragonet 



False Jacopever 



Hake 



20 30 40. 50. 60 70. 80. 90. 100. 

 Bray-Curtis similarity 



-H Cuttlefish 



Figure 6 



Inverse classification diagram showing species' associations among all the samples (pooled by major substratum and dive) collected 

 in the sites to the north and south of the Orange River mouth. 



that has focusseci on the commercially important demersal 

 fisheries along the west coast of southern Africa, there is a 

 dearth of published material that can be usefully compared 

 to the current observations. Roel ( 1987 ) recognized that the 

 species composition of the demersal fauna in the southern 

 Benguela ecosystem was mainly influenced by depth, and 

 she identified two distinct communities of demersal nekton. 

 One of these was dominated by Merluccius paradoxus and 

 occuired at depths greater than 380 m, and the other was 

 dominated by M. capensis and was confined to the continen- 

 tal shelf In the northern Benguela ecosystem, Macpherson 

 and Roel (1987) recognized five distinct demersal fish 

 communities. Two of these were distributed over the shelf 

 (and were subdivided by latitude), and the balance was es- 

 sentially depth related and ran parallel to the coastline, 

 offshore. Similar observations on the structure of the 

 demersal fish assemblages were made by Macpherson 

 and Gordoa ( 1992) and Mas-Riera et al. ( 1990), who noted 

 that latitudinal variations in community composition were 

 generally associated with the state of upwelling. The latter 

 authors postulated that the especially low level of diversity 

 of the demersal fish community in the northern Beguela 

 ecosystem (the region of our study) was due to the presence 

 of low-oxygen bottom water The "indicator" species for the 

 assemblage in their analysis included Merlucxius capensis 



and Sufflogobius biharbatus, which were also among the 

 common species observed in our study. 



Smale et al. (1993) observed three demersal communi- 

 ties over the Agulhas Bank (south coast of South Africa) 

 which were related to depth; inshore (<100 m), shelf (90- 

 190 m), and slope (>200 m). Although the species observed 

 in the our surveys were all conspicuous components of the 

 shelf communities identified by the these authors, they 

 were relatively rare (or absent) from the samples collected 

 by Smale et al. (1993). Unfortunately, the nature of these 

 authors' data (30-60 minute trawls) is such that they do 

 not provide any fine-scale information on species associa- 

 tions, and this effectively limits further comparison. 



There appears to be a difference in the community of 

 fishes from the northern and southern study sites (e.g. the 

 presence of dragonets in the south but not in the north). 

 Differences in the infaunal communities from undis- 

 turbed sites to the north and south of the Orange River 

 were noted by Field et al. ( 1996). Gibbons and Sulaiman'' 



Gibbons, M. J., and A. Sulaiman. 1998. A video-description of 

 the mid-shelf benthic environment off the west coast of south- 

 em Africa, with a comment on the habitat association of demer- 

 sal nekton. A report to De Beers Marine (Pty) Ltd., Cape Town 

 8000, Republic of South Africa, 62 p. 



