420 



Fishery Bulletin 94(3), 1996 



A possible reason why few red spot whiting escaped 

 continuously during trawling is that there were in- 

 sufficient stimuli generated in the codend to invoke 

 an escape response. This hypothesis is supported by 

 previous studies that have shown that physiological 

 and behavioral differences between species are ma- 

 jor factors in the effectiveness of escape panels 

 (Robertson and Stewart, 1988; Briggs, 1992; 

 Fonteyne and M'Rabet, 1992). The subsequent es- 

 cape of red spot whiting from the 85-mm-long codend 

 during the 10-15 s haulback delay may have been a 

 result of additional stimuli generated by changes in 

 the geometry of the codend. For example, Watson 

 ( 1989) observed that during haulback, differences in 

 the contours of the codend and associated water flow 

 caused fish to aggregate and become disorientated. 

 The fish then increased their swimming speeds and 

 randomly charged the meshes around the codend. 

 By opening the meshes in this area (i.e. by our inclu- 

 sion of the 85-mm-long square-mesh panel), we may 

 have stimulated some fish to escape the trawl. 



The delayed escape of red spot whiting from the 

 85-mm-long codend raises important questions con- 

 cerning the extent of their mortality during trawl- 

 ing and after escape. Because previous studies have 

 shown that fish that remain in the codend experi- 

 ence greater stress and fatigue than those that es- 

 cape continuously, those individuals escaping at the 

 end of the tow may suffer greater mortality. Further, 

 such post-trawl mortality, particularly among 

 smaller fish, may increase in proportion to tow du- 

 ration. Obviously such effects on red spot whiting 

 should be investigated prior to commercial use of the 

 85-mm-long codend per se, because, unless most of 



the escaping fish survive, such designs are of little 

 value. 



It may be possible to reduce the problems of trawl 

 mortality for red spot whiting by altering the design 

 of the 85-mm-long codend so that more fish escape 

 continuously during trawling. One possible modifi- 

 cation to the design would be to make the square- 

 mesh panel longer (see Briggs, 1992), allowing more 

 red spot whiting to escape at random along the 

 codend. This may also provide juveniles of other com- 

 mercially important species (i.e. cuttlefish, octopus, 

 and eastern blue spot flathead) the opportunity to 

 escape, although there was little evidence in our ex- 

 periments to suggest that these species displayed 

 active escape response toward the square-mesh panel 

 (see Figs. 2 and 3; Tables 1-3). Their lack of exclu- 

 sion is more likely attributable to intraspecific vari- 

 ability in behavioral responses rather than to the 

 design of the square-mesh panel (see also Robertson 

 and Stewart, 1988; Briggs, 1992). 



Another modification to allow more red spot whit- 

 ing to escape the trawl continuously may be to in- 

 crease the size of mesh used in the panels. However, 

 the size-frequency composition of red spot whiting 

 captured in the 85-mm-long codend and control 

 showed that the size of mesh used in the panel al- 

 lowed fish within the range of sizes sampled to es- 

 cape, including some individuals of commercial size 

 (>16 cm) (Fig. 4). Further, this size of mesh resulted 

 in no significant reduction in the weight of prawns, 

 nor were there any significant differences detected 

 in the size composition of prawns between the con- 

 trol and the 85-mm-long or 85-mm-wide codends. 

 Subsequent increases in the size of mesh may, there- 



