486 



Fishery Bulletin 100(3) 



North 



-^ 



-e 



HARD 



HARD 



HARD 



HARD 



MIXED 



HARD 



SOFT 



SOFT 



MIXED 



SOFT 



MIXED 



MIXED 



MIXED 



HARD 



MIXED 



HARD 



SOFT 



HARD 



HARD 



HARD 



HARD 



— SOFT 



— SOFT 



— HARD 



— MIXED 



— MIXED 

 -HARD 



— MIXED 



— MIXED 



— MIXED 



— SOFT 



— MIXED 



— SOFT 



— SOFT 



— SOFT 



— SOFT 



— MIXED 



— SOFT 



— SOFT 

 J- SOFT 

 P-SOFT 



H r MIXED 

 LrSOFT 

 '-SOFT 



rSOFT 



'-SOFT 



SOFT 



MIXED 

 MIXED 

 SOFT 

 MIXED 



30. 



40. 



50. 



60. 



70. 



80. 



90. 



100. 



3rav-Curti^ siniilantv 



South 



SOFT 



40. 50. 60. 70. 80. 

 Bray-Curtis similarily 



90. 100. 



Figure 5 



Cluster analysis of the pooled (by major substratum per dive I samples of nekton collected over the areas to the north and south of 

 the Orange River mouth. 



bycatch from the hake-directed trawl fishery and are thus 

 assumed to occupy a wide range of habitats. 



The gobies seen in our study tended to show neither a 

 strong association with any substratum type (and their 

 associated biota ), nor were they linked to areas with a rich 

 ophiuroid fauna (Tables 4 and 6). Sufflogobius bibarbatus 

 is also known as the pelagic goby because it spends much 

 of its time at night in the water column,- where it forms 

 a conspicuous part of the fish community'^ and where it 

 feeds largely on zooplankton.- If the gobies obsei-ved in 

 our study were S. bibarbatus (as postulated), then the 

 data suggest that their use of the demersal habitat by day 

 was primarily linked to shelter. Othei-wide, it might be 

 expected that they would be associated with areas rich in 



- Pillar, S. C. 2000. Personal commun. Marine and Coastal 



Management, Cape Town, South Africa. 

 •^ OToole, M. 2000. Personal commun. Ministry of Fisheries 



and Marine Resources, Windhoek, Namibia. 



benthic food, such as ophiuroids, which are known to form 

 part of the diet of other gobies (Gibson, 1982). 



Communities 



The overall diversity of the fish community obsei-ved 

 by the Jago is less than that determined from the same 

 region with trawl nets (Roel, 1987), principally because 

 the area sampled was significantly smaller. The diversity 

 of the fish and cephalopod fauna may also seem low by 

 comparison with submersible surveys elsewhere (e.g, 

 Felley et al., 1989). But, this region is not known to sup- 

 port a high diversity of demersal species (Mas-Riera et al., 

 1990) because of the perceived harshness of the prevailing 

 environment (such as frequent intrusions of low-oxygen 

 bottom water (Bailey and Rogers, 1997)). 



The demersal nekton fauna, both as individual species 

 and as a whole, was strongly influenced by substratum type 

 (as in Parker et al., 1994). Despite the large research effort 



