Hickford and Schiel: Gillnetting in southern New Zealand 



673 



600 

 500 



400 



300 



200- 



100- 



 Gilled 

 B Wedged 

 QTangled 



in w 



l- g 



>.: 



500 



400 



300 



ZOO 



100- 





t- O tr i < 



D 5 t- < LlJ 

 CD 5 O 2 -< 



ct t 

 W 1/5 



Species 



Figure 1 



The mean fork length ( + 1 SE) offish captured by each en- 

 tanglement mode in the (A) 2.5" mesh, (B) 3.5" mesh, and 

 (C) 4.5" mesh, n = the number offish from each species 

 caught in each method. See Table 3 for species codes. 



or the two set durations in the number of Aplo- 

 dactylus arctidens, Latridopsis ciliaris, or Arripis trutta 

 caught per hour (all F values were not significant). 



The proportion of damaged fish in the landed catch 

 was small for nets of all three mesh sizes set for six 

 hours but increased markedly for the longer set times 

 (Fig. 5). ANOVAof the condition index of 16 fish se- 

 lected randomly from each mesh size and each set 

 time showed that fish were significantly more dam- 

 aged in the longer sets (F 190 =19.23, P<0.001), but 

 there was no significant difference among the three 

 mesh sizes in the degree of damage suffered by landed 

 fish (F 290 =2.76, P=0.069). 



Maximum girth (mm) 



 Wedged OGilled n Tangled 



Figure 2 



Proportion of butterfish, Odax pullus, caught by 

 each entanglement mode in the (A) 2.5" mesh 

 and (B)3.5" mesh, n = the number of fish in each 

 size class. 



Discussion 



Each species showed a distinctive pattern in its form 

 of entanglement in the three mesh sizes. These pat- 

 terns appear to be a consequence of the behavioral 

 and morphological characteristics unique to each 

 species. For example, Arripis trutta were caught 

 mainly by being gilled in the nets. This species is a 

 pelagic carnivore that is dependent on a strong swim- 

 ming thrust for catching prey. Once gilled, they would 

 be expected to drive forward firmly into the net and 

 to become wedged. The low number of wedged fish 

 for this species may be a result of their firm flesh, 

 which is not easily compressed by the mesh and 

 which may prevent them from entering the net fur- 

 ther. Larger fish, despite their greater swimming 

 thrust (Lander, 1969), cannot enter the small mesh 

 sizes far enough to become wedged. 



Odax pullus were mostly gilled and wedged in the 

 nets. The low number of O. pullus that become 

 tangled was likely the result of the soft fin rays, fused 

 teeth, and small scales typical of this species (Paulin 

 et al., 1989), all of which offer little that will snag on 



