250 
Fishery Bulletin 99(2) 
Table II 
Resident catches (all species) in each of the twelve pots in the experimental fleet observed after initial deployment of gear at time 
intervals from August 1995 to September 1996. Mean resident catch per pot is also shown. “R” indicates those pots that were 
removed at each time interval 
Pot number 
Month 
Day 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
Mean ± 95% Cl 
Aug 
1 
1 
2 
0 
0 
1 
4 
4 
1 
4 
2 
2 
0 
1.75 ±0.87 
Aug 
4 
2 
2 
1 
0 
2 
4 
4 
1 
4 
3 
4 
1 
2.33 ±0.81 
Aug 
13 
3 
2 
2 
2 
3 
4 
4 
4 
6 
0 
4 
0 
2.83 ±0.99 
Aug 
27 
2 
2 
3 
3 
4 
7 
4 
6 
5 
1 
3 
2 
3.50 ±1.01 
Sep 
40 
0 
2 
3 
3 
4 
5 
4 
6 
5 
1 
3 
1 
3.08 ±1.04 
Oct 
69 
0 
2 
6 
3 
5 
7 
5 
6 
7 
1 
1 
3 
3.83 ±1.40 
Nov 
88 
0 
2 
9 
4 
5 
7 
3 
5 
7 
1 
1 
2 
3.83 ±1.60 
Nov 
101 
R 
3 
8 
R 
6 
6 
3 
R 
7 
1 
1 
2 
4.11 ±1.74 
Dec 
125 
R 
3 
5 
R 
6 
2 
5 
R 
7 
2 
2 
1 
3.67 ±1.39 
May 
270 
R 
1 
0 
R 
5 
2 
5 
R 
2 
0 
0 
0 
1.67 ±1.65 
Jul 
333 
R 
5 
2 
R 
3 
R 
4 
R 
4 
0 
R 
R 
3.00 ±1.17 
Aug 
369 
R 
7 
0 
R 
2 
R 
5 
R 
3 
2 
R 
R 
3.17 ±1.62 
Sep 
398 
R 
7 
0 
R 
2 
R 
5 
R 
3 
2 
R 
R 
3.17 ±1.62 
enough to pass through the meshes of the pots. Thus we 
were able to quantify the minimum mortality of animals 
captured per pot per year for the data up to the first 369 
days (approximately 12 months) after initial deployment. 
The effect of substratum type on the movement of pots 
on the seabed was analyzed for the first 7 observation pe- 
riods by using one-way ANOVA. Data from the remaining 
observation periods were not used because some of the 
pots were removed after the seventh sampling date. Three 
different substratum categories were used: rock and large 
boulders; medium boulders and cobbles; and mixed sedi- 
ments. 
Results 
The bait within the pots remained secure during their 
deployment. The bait was consumed rapidly, and by day 
13 only remnants of the bait were observed in 9 of the 
pots, and by day 27 only 1 pot had any remnants of 
bait. Throughout the experiment , none of the pots became 
detached from the mainline between the sinkers. Pot 
movement was greatest in the first few weeks after deploy- 
ment and was related to seabed topography. The entire 
fleet of pots tended to move in a north-westerly direction 
in line with the prevailing tidal currents as determined 
from the position of the surface marker buoys. Substratum 
type did not affect the extent to which the individual pots 
moved on average during the first 88 days of the experi- 
ment (ANOVA, i\5 9 =0. 6 2, P=0.56). 
Throughout the experiment seven different species were 
captured in the pots. Divers observed that some animals 
were always resident in the pots because previously ob- 
served animals were frequently observed on subsequent 
sampling dates. The number of animals resident in the 
pots did not differ significantly between different observa- 
tion periods (Table 1, GLM F V2 112 =1.07,P=0.39).This find- 
ing suggests that as animals died or escaped, they were re- 
placed by new catches. Only spider and brown crabs were 
captured in sufficient numbers for more detailed analyses. 
Analyses of the CPUE up to 125 days after initial deploy- 
ment of the pots revealed that the CPUE decreased sig- 
nificantly for spider crab and brown crab (Fig. 2). The best 
relationship between the decline in CPUE with time was 
given by the following relationships: 
CPUE spider crab = 
0.0216 + (0.377/day) [r=0.97, F l 7 =236.7, P<0.0001], 
CPUE broum crab - 
0.0205 + (0.244/day) [/-0.92, P 17 =77.5, PcO.0001]. 
It is clear from Figure 2 that the CPUE for spider crab and 
brown crab declined to a minimum rate between 125 and 
270 days after initial deployment and then increased rap- 
idly before decreasing again. The regression equation used 
to calculate the total predicted catch for the first 125 days 
of the experiment gives a catch of 4.74 spider crabs and 
3.88 brown crabs per pot. 
Catch rate, however, does not provide an indication of 
the total mortality of animals associated with the ghost- 
fishing pots. We were able to confirm annual mortality on- 
ly per pot for lobster and brown crabs, of which 100% died 
because divers retrieved all the tags deployed on individu- 
al lobsters and brown crabs in the pots (Table 1). A conser- 
vative estimate of total mortality indicates that each pot 
killed on average 6.06 brown crabs and 0.44 lobsters per 
year (Table 2). This finding takes no account of animals 
