224 
Fishery Bulletin 115(2) 
More on wire 
More on wire 
swo 
Billfish 
Tuna- 
DOL- 
-•j- ^ 
•1 B 
1 
1 
—1«- 
1 
f 
1 
1 
»•- 
1 —•— 
1 
1 _ 
1 ■ 
1 • 
' m . 
• 
1 • 
1 
1.0 1.5 2.0 
Odds-ratio 
2.5 
2 4 6 
Odds-ratio 
Figure 3 
Odds-ratios of the (A) catch rates in number and (B) hooking mortality 
for the change from the leader material from monofilament to wire for 
species and species groups caught during longline sets conducted during 
November 2013-March 2014 in the southwest Indian Ocean. Food and 
Agriculture Organization of the United Nations species codes are used: 
SWO^swordfish (Xiphias gladius); DOL=dolphinfish (Coryphaena hippu- 
rus); BSH^blue shark {Prionace glauca); GES=snake mackerel {Gempylus 
serpens); ALX=longnose lancetfish (Alepisaurus ferox); and LEC=escolar 
(Lepidocybium flavobrunneum). The models and odds-ratios were calculat¬ 
ed only for species or combined taxa with 30 or more individuals. The er¬ 
ror bars indicate the 95% confidence intervals of the odds-ratios. The ver¬ 
tical dashed line indicates the reference of the odds-ratios (l=no change). 
The top arrow represents the increase in the odds-ratios when changing 
from 2.5-mm monofilament nylon to 1.2-mm multifilament stainless wire 
leaders. 
catch rates was compensated by the increase of blue 
shark retention (Table 1). 
The mean bite-off rates for wire and monofilament 
leaders were 1.38/1000 hooks and 5.37/1000 hooks, re¬ 
spectively. These differences were found to be significant 
(permutation test: difference in means=3.99, P<0.001). 
Regarding hooking location, most shark species were 
deeply hooked (58% overall, 60% for the blue shark 
and 50% for the shortfin make (Isurus oxyrinchus), 
respectively). The pelagic stingray was an exception, 
where specimens were mostly hooked by the mouth or 
jaw (83%). In contrast, most tunas and billfish species 
were retained, having been hooked in the mouth, ex¬ 
cept for the swordfish, which was predominantly deeply 
hooked (85%, and 12% and 2% were retained, having 
been hooked in the mouth and externally, respectively). 
No differences were detected when comparing hooking 
location between the 2 different types of leader mate¬ 
rial, for the main species, the swordfish (contingency 
table analysis: P>0.05 for all cases; Fig. 4) 
MoitaSity rates 
The effects of the leader type in terms 
of mortality rates (in number) at haul- 
back varied among the different spe¬ 
cies groups and taxa. The mortality 
models detected significantly less mor¬ 
tality from hooking for the swordfish 
when wire leaders were used, show¬ 
ing a 16% decrease in the mortalities 
(95% Cl varying between 2% and 28%) 
and also for the billfishes group with a 
15% decrease in the mortalities (95% 
Cl varying between 1% and 26%). In 
contrast, for the other bony fishes 
there was an increase in hooking 
mortality of 55% when wire leaders 
were used (95% Cl varying between 
17% and 100%). For the blue shark 
the hooking mortality also increased 
when wire leaders were used (26% in 
the point estimate) but this increase 
was not statistically significant be¬ 
cause the 95% CIs varied between a 
decrease of 6.5% and an increase of 
71% with the use of wire. The same 
was observed for the shark group com¬ 
bined; the point estimate showed a 
decrease in hooking mortality of 27% 
when wire leaders were used, but the 
95% Cl varied between a decrease of 
5% and an increase of 70%, indicating 
that the effects were not statistically 
significant (Tables 1 and 2, Fig. 3). 
Size distribution of retained species 
Statistics of the size structure of the 
species caught on the different leader 
materials tested during this study are 
summarized in Table 3. For most species caught, there 
were no major differences in size range and mean size 
for the different leader types tested (Table 3). However, 
for the blue shark, a wider size range was recorded and 
significant differences in the mean size were detected, 
and slightly larger sizes were captured with wire lead¬ 
ers (mean size: 198.2 cm FL [SD 38.15] vs. than with 
monofilament leaders (mean size 190.6 cm FL [SD 
35.32]). For swordfish, very similar size ranges were 
observed and no significant differences were detected 
in the mean size captured for the 2 leader materials 
(Fig. 5, Table 3). 
Discussion 
The results from this study show that leader material 
had an effect on the catch composition of a longline 
fishery and that wire leaders caught more species. This 
result contrasts with those reported by Ward et al. 
