BULLETIN OF MARINE SCIENCE. 88(3):647–665. 2012
http://dx.doi.org/10.5343/bms.2011.1086
Bulletin of Marine Science 647
© 2012 Rosenstiel School of Marine & Atmospheric Science of
the University of Miami
EFFECT OF CIRCLE HOOK SIZE ON REEF FISH CATCH
RATEs, SPECIES COMPOSITION, AND SELECTIVITY IN THE
NORTHERN GULF OF MEXICO RECREATIONAL FISHERY
William F Patterson III, Clay E Porch,
Joseph H Tarnecki, and Andrew J Strelcheck
ABSTRACT
The effect of circle hook size on reef fish catch rates, species composition, and
selectivity was tested in the northern Gulf of Mexico recreational fishery. Fish
communities first were sampled at natural (n = 19) and artificial (n = 23) reefs
with a micro remotely operated vehicle (ROV) equipped with a laser scale. Fishing
experiments (n = 69) then were conducted with 9/0, 12/0, and 15/0 circle hooks.
Hook size significantly affected fish catch rates, species composition, and size. Small
invertivore fishes constituted 33.4% of the catch taken with 9/0 hooks, but were
nearly absent from the catch made with larger hook sizes. In contrast, red snapper,
Lutjanus campechanus (Poey, 1860), constituted only 25.3% of fishery species
total abundance in video samples, but ranged from 59.1% of the 9/0 hook catch to
nearly 90% for 15/0 hooks. A novel maximum likelihood approach was developed
to estimate exponential-logistic selectivity functions for each hook size from ROVbased
estimates of red snapper size distributions and observed hook-specific catch
at size. Both the 9/0 and 12/0 hooks displayed dome-shaped selectivity functions,
while the 15/0 hook size was estimated to have a logistic-shaped function. However,
observed catch-at-size data displayed a dome-shaped pattern for 15/0 catches when
paired with 9/0 hooks, but an indistinct pattern when paired with 12/0 hooks.
Overall, study results suggest that regulating circle hook size would affect reef
fish catch rates and size in the northern Gulf of Mexico recreational fishery, but
potential conservation benefits may be confounded by unintended effects.
Bycatch of undersized fishes and non-targeted species is a global fisheries problem
(Alverson et al. 1994, Lewison et al. 2004, Kelleher 2005, Davies et al. 2009). Circle
hooks have been proposed as a conservation measure to reduce bycatch rates or mitigate
issues related to incidental hooking of non-targeted species. The scientific literature
on circle hooks consists predominantly of studies in which their conservation
benefits have been tested in commercial fisheries targeting oceanic pelagic species
(Kaplan et al. 2007, Carruthers et al. 2009, Sales et al. 2010, Pacheo et al. 2011), and
such studies are well-presented in this volume (Serafy et al. 2012). However, circle
hooks also may have significant conservation benefits for reducing the incidental
catch of undersized and non-targeted fishes in recreational hook-and-line fisheries
(Grover et al. 2002, Prince et al. 2002, Bacheler and Buckel 2004).
Recreational fishing effort has a greater impact on marine fisheries off the southeastern
United States than in any other region of the US, both in terms of total
landings, as well as landings of stocks estimated to be overfished or undergoing overfishing
(Coleman et al. 2004, Hanson and Sauls 2011, Ihde et al. 2011). Increasingly, it
is not just the landed catch that is at issue, but also dead discards of undersized, nontargeted,
or closed-season fishes that contribute significantly to the total harvest for
OA
Open access content
648 BULLETIN OF MARINE SCIENCE. VOL 88, NO 3. 2012
several marine species. Red snapper, Lutjanus campechanus (Poey, 1860), and gag,
Mycteroperca microlepis (Goode and Bean, 1879), are two of the more highly targeted
marine fishes in the US Gulf of Mexico (GOM), and management of both, as well as
the reef fish fishery in general, is greatly affected by bycatch issues (Goodyear 1995,
Johnson et al. 1997, Strelcheck and Hood 2007). Recent stock assessments have highlighted
the fact that dead discards in the recreational fishery contribute a substantial
percentage (i.e., 20%–30%) of the total recreational harvest of red snapper and
gag, both of which have a spawning stock biomass estimated to be severely depleted
(Porch 2007, SEDAR 2006, 2009a,b).
There are several aspects of the biology of northern GOM reef fishes, as well as
their fisheries management, that confound issues related to discarding. Fishery managers
have relied mostly on traditional management tools, such as size limits, daily
bag limits, and closed seasons, in attempts to limit recreational effort, hence fishing
mortality (F). However, the GOM reef fish fishery is diverse, with 42 species in the
Gulf of Mexico Fishery Management Council’s (GMFMC) reef fish fishery management
plan. Catch rates for fishes such as red snapper and gag can be high even during
closed seasons given the non-specificity of fishing effort when recreational seasons
are open for other reef fishes. Furthermore, many of the snappers and groupers in
the northern GOM display ontogenetic movements into deeper waters, being found
out toward the shelf edge (approximately 200 m) or deeper as adults (Mitchell et
al. 2004, Lindberg et al. 2006, Patterson 2007), where water depth can exacerbate
discard mortality. Venting is required for regulatory discards in the GOM reef fish
fishery, but acute and chronic barotrauma severely affects physoclistous fishes that
experience rapid pressure changes of several atmospheres when brought from depth,
which in turn translates into high post-release mortality rates even for fish that successfully
return to depth after being released (Gitschlag and Renault 1994, Wilson
and Burns 1996, Rummer 2007).
The GMFMC has discussed employing non-traditional management tools, such
as tag programs, limits on days-at-sea for for-hire vessels, marine protected areas,
and catch shares, to reduce recreational F in the GOM reef fish fishery, partly because
the pervasive issue of discard mortality hampers the utility of many of the
traditional tools the GMFMC has relied upon. One potential method to reduce catch
rates, especially for undersized fishes, is to regulate the selectivity of the hook-andline
gear used to target reef fishes. Circle hooks have been required in the GOM reef
fishery since 2007, but the size of hooks that may be used is not currently regulated
(GMFMC 2007). The GMFMC considered regulating hook sizes for harvesting reef
fishes but opted not to implement hook size restrictions due to limited scientific
research pertaining to reef fishes and a lack of standardization in hook sizes among
manufacturers (GMFMC 2007). The goal of the present study was to examine the
effect of circle hook size on reef fish catch rates, species composition, and size distributions
in the GOM recreational reef fish fishery. Our approach was to sample reef
sites in the northern GOM with a micro remotely operated vehicle (ROV) to estimate
community and size structure of reef fishes present on reefs. Then, we fished the sites
with different sized circle hooks to test the effect of hook size on catch rates, species
composition, and size distributions. Lastly, we developed a novel method to estimate
hook selectivity for red snapper from estimated in situ size distributions and size
distributions of hook-specific catches.
PATTERSON III ET AL.: CIRCLE HOOK SIZE EFFECT IN REEF FISH FISHERY 649
Methods
Field Sampling of Reef Fishes.—Sampling occurred at natural (n = 19) and artificial reef
(n = 23) sites located at depths between 20 and 67 m on the northeastern Gulf of Mexico continental
shelf (latitude range: 85.92N–88.17N, longitude range: 29.50W–30.20W) between
June 2009 and August 2010. Video samples (n = 69) of fish communities were obtained with
either a VideoRay Pro3 (dimensions: 30 cm long, 24 cm tall, 22 cm wide; mass = 3.8 kg) or
Pro4 micro ROV (dimensions: 36 cm long, 28 cm tall, 22 cm wide; mass = 4.8 kg). Both ROVs
have a depth rating of 170 m, a wide angle (105° or 116°, respectively) lens on a 570-line forward-
looking color camera, and were equipped with a red laser scale (10 cm between lasers)
to estimate fish size (Patterson et al. 2009). The ROVs were tethered to the surface where
they were controlled by a pilot via an integrated control box containing a 38-cm monitor to
observe video captured by the ROV’s camera during sampling.
Video sampling was conducted at study reefs either with the point-count method described
by Patterson et al. (2009) or a transect method, depending on habitat type and dimensions.
The point-count method was used to sample a 15-m wide cylinder around isolated reef habitat,
such as single artificial reef modules, while the transect sampling method was utilized
for reef habitat that was more broadly distributed, which was characteristic of natural reef
habitat examined in the present study. Transects were flown approximately 2 m above the sea
floor and the camera angle adjusted such that the field of view was approximately 10 m wide.
Orthogonal 25-m long transects were flown from a center point at a given reef site such that
approximately 1000 m2 of reef habitat were sampled.
Analysis of video samples was performed with a Sony DVCAM DSR-11 digital VCR capable
of frame by frame playback and a Sony LMD-170 high resolution LCD monitor. When
the point-count method was employed, fish counts were summed among all sampling segments
and then divided by the sampling cylinder’s area (176.7 m2) to estimate fish density (see
Patterson et al. 2009). Fish density was computed by summing taxa-specific fish counts and
then dividing by the total area among transects. The mean density of fishery species among all
samples was computed, and then the relative proportion of total fishery species density was
computed for each species.
Fork length (FL) was estimated for fishes struck by the laser scale during video sampling
at study reefs. For a given fish, this was accomplished by multiplying its length measured in a
video frame by the known distance between lasers (100 mm), and then dividing that product
by the distance measured between lasers in the frame. FL was converted to total length (TL)
with regression equations developed from fish captured with hook and line. Bias-correction
in fish length estimates then was conducted based on results from a pool experiment in which
model fish length was estimated for different angles from perpendicular and distances from
the ROV (Patterson et al. 2009). For conditions observed in situ, the mean bias of underestimating
fish length was estimated to be 3% with a standard deviation of 0.6%. Therefore, TL
estimates were adjusted based on a random probability and normally distributed bias with
mean equal to 3% and standard deviation equal to 0.6%.
Red snapper age distribution was estimated from length with an age-length key for laserscaled
fish as well as fish captured with circle hooks. The key was computed from size-at-age
data (n = 1755) reported in Patterson et al. (2001) for fish from the north central GOM that
were aged via analysis of sagittal otolith thin sections. Size-at-age data from additional red
snapper samples (n = 465) were added to the data set from fish sampled in the north central
GOM off northwest Florida and Alabama in 2009–2010 (WF Patterson, unpubl data). The
probability that fish within 10 mm length bins in the combined size at age data set were a
given age was computed, and then age was assigned probabilistically to bias-corrected lengths
of laser-scaled and hook and line sampled fish.
Fish were captured with hook and line for 30 min by 6–8 anglers at each reef site following
video sampling. Two-hook bottom rigs were deployed on each fishing rod and consisted
of a 1.5-m leader constructed of 60-lb (approximately 27.2 kg) monofilament which had two
650 BULLETIN OF MARINE SCIENCE. VOL 88, NO 3. 2012
shorter leaders extending approximately 0.5 m horizontally from the main leader. Terminal
tackle on the ends of the horizontal leaders was either 9/0, 12/0, or 15/0 Mustad 39660D circle
hooks (Fig. 1). This range in hook sizes was selected due to the fact that 9/0 Mustad 39660D
circle hooks are most similar in size to 3/0 J-hooks traditionally used in the GOM reef fish
fishery and 15/0 hooks are the upper limit in hook size typically employed by recreational
anglers. Hooks were fished with either cut squid or mackerel scad, Decapterus macarellus
(Cuvier, 1833), as bait. At a given site, half the anglers fished with one size of circle hooks
and the other half fished with another size (combination-1 = 9/0 and 12/0 hooks, combination-
2 = 9/0 and 15/0 hooks, and combination-3 = 12/0 and 15/0 hooks), although during 10
sampling events, only one hook size was used. Captured fish were either randomly sampled
and retained to provide biological samples for another study, or were returned to the water
following length measurement and venting their gas bladders.
Statistical Analysis.—The difference in catch rate among hook sizes was tested with
one-way analysis of variance (ANOVA, α = 0.05) for all fishes and separately for red snapper.
Catch per hook hour was log-transformed to meet the assumption of normality prior to
analysis in SAS (SAS, Inc. 1998). Differences among the relative proportion of fishery species
estimated among reef sites, and standardized to sample area, and the proportion of those species
in the hook-specific catches were tested with a one-way analysis of similarity (ANOSIM,
α = 0.005) model in the Primer software package (Clark and Gorley 2001). Species-specific
proportions were square-root transformed and standardized prior to analysis. The difference
in size of fish captured among hook sizes was tested with ANOVA (α = 0.05) following logtransformation
to meet parametric assumptions.
Red snapper sample sizes were thought to be large enough among laser-scaled individuals
and hook-specific catches that selectivity could be estimated directly for this species. In
many cases, reef sites were limited in their spatial extent such that it was possible to catch a
significant fraction of the local red snapper population. Assuming deaths by natural causes
were a negligible source of mortality during the short fishing event, an appropriate model for
the catches and ROV observations is:
Figure 1. Digital image of 9/0, 12/0, and 15/0 Mustad 39660D circle hooks used to test the effect
of hook size on reef fish catch rate and size distribution in the northern Gulf of Mexico recreational
reef fish fishery.
PATTERSON III ET AL.: CIRCLE HOOK SIZE EFFECT IN REEF FISH FISHERY 651
C F
f q S N e
V edN
F fq S
1
lhk
l
hk h lh lk
F
lk lk
lk hk h lh
h
lk
=
-
=
=
Z ^ - h
[
\
]]
]]
/ (Eq. 1)
where Nlk is the number of red snapper of length l at fishing location k, Clhk is the number of
red snapper caught by hook type h, and Vlk is the number of red snapper measured during the
ROV survey. The variable F represents the total fishing mortality rate in the area, which is the
sum of the fishing mortality rates associated with each of the two hook types being fished
simultaneously. The variables f and e represent the relative effort expended during fishing or
the ROV survey. The variables q and d represent the relative fishing power of each hook type
and relative detectability of red snapper during the ROV survey. The variable S represents the
selection function (see below).
The C and V are observed quantities and the effort parameters are controlled with little
error, leaving the q, d, S, and N to be estimated. If the observations C and V were mutually
exclusive outcomes, then the estimation could be accomplished using the SELECT model of
Millar (1992), which would conveniently eliminate the need to estimate the nuisance parameters
(N). However, since the survey was conducted prior to fishing, it is likely that some of
the fish surveyed were also caught and retained. Alternatively, if the size distribution of fish
in the surveyed area (V) is measured with little error, then the system of equations in (1) may
be reformulated as
C edF
f q S V e
F fq S
1
lhk
lk
hk h lh lk
F
lk hk h lh
h
lk
=
-
=
Z ^ - h
[
\
]]
] / . (Eq. 2)
Assuming the total red snapper catch by each hook type at each location (Chk) has approximately
a normal distribution (with variance s2) and that the proportion of the catch falling
in each length category (plhk = Clhk / Chk) has approximately a multinomial distribution, then
maximum likelihood estimates of q, d and S may be obtained by minimizing the negative loglikelihood
expression
L . log
c c
0 5 log n p p
2
hk 2
obs
hk
h,k e h,k h,k lhk
obs
= v v l e lhk
-
/ a k - +/ / . (Eq. 3)
The superscript obs distinguishes the observed data from the value predicted by the model
and n indicates the effective sample size (which in some cases might not equal the total catch).
The detectability parameter d in Equation 2 is confounded with the catchability parameters
q unless additional information is provided to estimate F, such as might be obtained if a
second visual survey were conducted after fishing occurred. In the present study, d was fixed
to 0.1 because approximately 10% of the red snapper in each area were able to be measured.
The exact value is inconsequential because only the relative fishing power of the gears is of
interest here.
The magnitude of the selection vector Slh is similarly confounded with the value of qh.
Moreover, the model can become over-parameterized if unique selection values are estimated
for each length category. A common solution to these two problems has been to model selection
as a mathematical function of length. Little has been published to guide the choice of
hook selection models; however, an inspection of the data indicated the 9/0 hooks caught
a lower proportion of large fish than the two larger hook sizes. This implies that the hooks
used in our study might exhibit an asymmetric, dome-shaped selection pattern. Two possible
candidates that were examined here are the exponential-logistic and double logistic curves:
652 BULLETIN OF MARINE SCIENCE. VOL 88, NO 3. 2012
S
e
e
e
e
double
1 1
1
1 1 1
exponential–logistic
logistic
1
l
l
l
l
1
2
b
=
- -
+
- +
a i
ba i
a i
b i
-
-
- -
- -
^
^
^
^
^
^
h
h
h
h
h
h
Z
[
\
]]
]]
(Eq. 4)
where α, β, q, q1, and q2 are parameters to be estimated and l is the midpoint of size interval l.
Note that both functions tend toward a flat-topped logistic function as β tends toward zero.
The data were pooled across all locations fished by the same combination of hook types due
to the relatively sparse number of video measurements in each unique ROV sample. Thus,
there were effectively three locations in terms of Equations 2 and 3, those fished in tandem by
9/0 and 12/0 hooks, 9/0 and 15/0 hooks, or 12/0 and 15/0 hooks. Bias-corrected laser-scaled
red snapper TL estimates were binned into 20 length categories defined by the boundaries:
170, 230, 250, 270, 290, 310, 330, 350, 370, 390, 410, 430, 450, 470, 490, 510, 530, 570, 610,
650, and 730 mm. Broader intervals were specified for the smallest and largest categories
because of the rarity of fish in those size classes in our samples. The initial model assumed the
same parameter values for the selection curves associated with each hook type (i.e., that there
was no hook effect) and that the shape of the selection curve was flat-topped (β near zero).
Additional parameters were estimated in a stepwise approach to allow for dome-shaped (β >
0) selection and hook-size effects. Akaike’s information criteria for small samples was used to
determine the combination of selection parameters that provided the most parsimonious fit
to the data (Hurvich and Tsai 1995).
Results
Fish counts from video analysis of ROV samples totaled 18,347 individuals belonging
to 108 taxa (94% of individuals were identified to species) among 69 samples.
Eleven fishery species accounted for 82% of the total fish count (see Table 1 for names
and authorities), with tomtate alone accounting for 34%. There was a distinct shift in
the relative proportion of those 11 species in hook-specific catches vs the fish communities
observed in video samples. For example, tomtate constituted 58.2% of the
total abundance among fishery species, yet was only 6.0% of the 9/0 catch and only
one tomtate was captured with a 15/0 hook. Red snapper, on the other hand, constituted
only 25.3% of fishery species observed in video samples, yet its percent abundance
in catches ranged from 59.1% for 9/0 hooks to nearly 90% for 15/0 hooks.
Catch rates were significantly affected by hook size for all fishes (ANOVA: F2,66 =
15.1, P < 0.001), but not for red snapper alone (ANOVA: F2,66 = 1.57, P = 0.215; Fig.
2). Overall, the size of captured fish was significantly affected by hook size (ANOVA:
F3,3061 = 42.1, P < 0.001; Fig. 3). Hook size also significantly affected the size of captured
red snapper (ANOVA: F3,1724 = 67.4, P < 0.001), other snappers (ANOVA: F3,402
= 6.65, P < 0.001), and groupers (ANOVA: F3,175 = 18.4, P < 0.001), but was not significant
for red porgy (ANOVA: F2,113 = 0.21, P = 0.809) or tomtate (ANOVA: F2,204 =
1.87, P = 0.157). However, sample sizes were low for red porgy and tomtate, both of
which had essentially no catch for 15/0 hooks, and most individuals caught of those
two species were near the median size observed in video samples regardless of hook
size (Fig. 3).
Cumulative frequency distributions of estimated red snapper size and age demonstrate
clear shifts between video samples and catches, with increasing trends in both
fish size and age with hook size (Fig. 4). Median estimated TL was 373 mm for red
PATTERSON III ET AL.: CIRCLE HOOK SIZE EFFECT IN REEF FISH FISHERY 653
Table 1. Estimates of percent relative abundance of fishery-important reef fish species observed in remotely operated vehicle collected video samples of fish
communities at natural and artificial reefs in the northern Gulf of Mexico, as well as percent catch of those species captured with 9/0, 12/0, and 15/0 Mustad
circle hooks.
Species Common name
Abundance among fishery species
in video samples (%) 9/0 catch (%) 12/0 catch (%) 15/0 catch (%)
Haemulon aurolineatum (Cuvier, 1830) Tomtate 58.2 6.0 1.8 0.5
Lutjanus campechanus (Poey, 1860) Red snapper 25.3 59.1 76.8 88.9
Pagrus pagrus (Linnaeus, 1758) Red porgy 7.5 17.5 9.7 2.3
Lutjanus griseus (Linnaeus, 1758) Gray snapper 3.4 0.9 0.0 0.5
Seriola dumerili (Risso, 1810) Greater amberjack 1.9 0.5 1.8 0.5
Lutjanus synagris (Linnaeus, 1758) Lane snapper 1.0 2.1 0.2 0.0
Balistes capriscus (Gmelin, 1789) Gray triggerfish 0.9 1.2 0.5 0.0
Rhomboplites aurorubens (Cuvier, 1829) Vermilion snapper 0.8 9.0 4.7 1.4
Mycteroperca phenax (Jordan and Swain, 1884) Scamp 0.7 2.5 2.0 1.9
Mycteroperca microlepis (Goode and Bean, 1879) Gag 0.1 0.5 1.4 1.9
Epinephelus morio (Valenciennes, 1828) Red grouper 0.1 0.7 1.1 2.3
654 BULLETIN OF MARINE SCIENCE. VOL 88, NO 3. 2012
Figure 2. Mean (+ SE) catch rate per hook hour for all fishes and red snapper, Lutjanus campechanus,
alone that were captured with 9/0, 12/0, and 15/0 circle hooks at reef sites in the northern
Gulf of Mexico. * Catch rate of all fishes declined significantly (P < 0.001) with increasing hook
size.
Figure 3. Boxplots of bias-corrected total length (TL) estimates of reef fishes scaled with a red
laser scale during remotely operated vehicle sampling of reef fish communities in northern Gulf
of Mexico and of measured TL of fishes that were captured with 9/0, 12/0, and 15/0 circle hooks.
Midline of each box indicates sample median. Lower and upper sides of each box indicate 25th
and 75th distribution percentiles, respectively. Extended bars represent 10th and 90th percentiles,
and filled circles indicate 5th and 95th percentiles. * P < 0.001.
PATTERSON III ET AL.: CIRCLE HOOK SIZE EFFECT IN REEF FISH FISHERY 655
snapper scaled with the laser scale attached to the ROV, and 424, 443, and 482 mm
for fish captured with 9/0, 12/0, and 15/0 hooks, respectively. Median estimated age
displayed a similar pattern in that it was 2.5 yrs for fishes scaled with lasers, and was
2.9, 3.2, and 3.6 yrs, respectively, for fishes captured with 9/0, 12/0, and 15/0 hooks.
However, the distribution of size or age in the sampled population is as critical as the
distribution of the catch when estimating selectivity, and in the present study there
were some differences in the size structure estimated among red snapper sampled at
sites fished with different hook size combinations (Fig. 5). Therefore, comparisons of
the overall size distribution of red snapper to the overall distributions of fishes captured
with the three circle hook sizes fished in our study are insufficient to evaluate
selectivity differences among hook sizes.
The double logistic model did not provide as good a fit to the data as the exponential
logistic, despite having one more parameter, therefore was not explored further.
Figure 4. Cumulative frequency distributions of red snapper, Lutjanus campechanus, (A) total
length and (B) age estimated with an age-length key for fish scaled with a red laser scale (n = 723)
during remotely operated vehicle (ROV) sampling of reef sites in the northern Gulf of Mexico, as
well as fish captured with 9/0 (n = 430), 12/0 (n = 314), and 15/0 (n = 174) circle hooks following
ROV sampling. Length estimates for laser-scaled fish were bias-corrected as described in the text.
656 BULLETIN OF MARINE SCIENCE. VOL 88, NO 3. 2012
All of the parameters of the exponential model significantly improved the fit to the
data except the β parameter for the 15/0 hook model (β15/0 = 0.002), which did not
differ significantly from zero, thus was not included in the final model (Table 2).
Resulting selectivity function parameters were highly correlated within hook types,
as is typical when a functional relationship is used, but not among hook types (Table
2). The general shapes of the selectivity functions estimated for 9/0 and 12/0 hooks
were dome-shaped, while the shape of the function estimated for 15/0 hooks was
logistic (Fig. 6). Fits of the exponential-logistic hook-specific selectivity models to
the red snapper catch data were reasonably good (Fig. 7). There are some noticeable
trends in the residuals of the fits (Fig. 7), but they are not consistent across locations.
Discussion
Our results indicate that circle hook size had a clear effect on reef fish catch rates,
species composition, and size distributions. The GMFMC mandated the use of circle
hooks in the GOM reef fish fishery in 2008 with Amendment 27 to its Reef Fish
Fishery Management Plan (GMFMC 2007), but the only stipulation is that hooks
Figure 5. Bias-corrected red snapper, Lutjanus campechanus, total length distributions estimated
with a red laser scale during remotely operated vehicle video sampling of reef sites in the northern
Gulf of Mexico. Combination-1 indicates sites where 9/0 and 12/0 circle hooks were fished.
Combination-2 indicates sites where 9/0 and 15/0 hooks were fished, and combination-3 indicates
sites where 12/0 and 15/0 hooks were fished.
PATTERSON III ET AL.: CIRCLE HOOK SIZE EFFECT IN REEF FISH FISHERY 657
Table 2. Maximum likelihood estimates (MLE) of parameters, coefficients of variation (CV, standard deviation/MLE), and correlation matrix for the final
exponential-logistic model fitted to circle hook catch-at-size data to estimate selectivity. In the table, q = catchability, ω = coefficient of variation of the observed
catch, α, β, and θ = exponential-logistic model parameters, and 9/0, 12/0, and 15/0 = circle hook sizes. Note that β15/0 did not differ significantly from the null
hypothesis (P < 0.001), thus was not estimated in the final model.
Correlation matrix
Parameter MLE CV q9/0 q12/0 q15/0 α9/0 θ9/0 β9/0 α12/0 θ12/0 β12/0 α15/0 θ15/0 ω
q9/0 0.163 0.121 1.00
q12/0 0.134 0.131 0.03 1.00
q15/0 0.059 0.106 0.04 0.03 1.00
α9/0 0.034 0.086 0.13 0.00 0.00 1.00
θ9/0 442.3 0.018 0.09 0.02 0.02 −0.79 1.00
β9/0 0.180 0.251 0.33 0.02 0.00 −0.42 0.43 1.00
α12/0 0.026 0.091 0.03 0.30 0.00 0.01 0.00 0.01 1.00
q12/0 486.4 0.028 −0.02 0.22 0.01 −0.03 0.04 0.00 −0.59 1.00
β12/0 0.300 0.322 0.00 0.42 0.00 −0.02 0.02 0.01 −0.17 0.64 1.00
α15/0 0.057 0.253 0.01 0.00 −0.18 0.01 −0.01 0.00 0.00 −0.01 0.00 1.00
q15/0 515.3 0.082 0.00 0.00 0.22 −0.01 0.02 0.00 0.00 0.01 0.00 −0.99 1.00
ω 0.136 0.299 0.01 −0.01 −0.04 0.03 −0.08 0.00 −0.01 −0.12 0.02 0.01 −0.02 1.00
658 BULLETIN OF MARINE SCIENCE. VOL 88, NO 3. 2012
cannot be made of stainless steel. Others have examined the conservation benefits of
circle vs J-shaped hooks, including in recreational hook and line fisheries, but results
from our study indicate that not only hook style but also hook size can have significant
effects. One goal of our study was to test whether using larger hooks would
reduce the catch of undersized fish, with red snapper (GOM recreational size limit
currently 406 mm TL) being a focus given its predominance among northern GOM
reefs and the fact that it is perhaps the most targeted fish in the reef fish fishery.
Using larger hooks did reduce the percentage of undersized red snapper caught, but
species diversity of the catch also decreased substantially as hook size increased,
with red snapper constituting nearly 90% of the catch taken with 15/0 circle hooks.
Therefore, if a management decision were made to require only larger circle hooks
to be used in the GOM recreational reef fish fishery, then a tradeoff for the benefit of
minimizing sublegal red snapper discards would be diminished catch rates of other
species. Furthermore, during the red snapper closed season, a minimum hook size
may actually increase the number of red snapper discards. This would likely occur
when fisherman sought other species that constituted a smaller percentage of the
total catch when fishing with larger hooks. Therefore, an increasing number of red
snapper would be discarded while seeking to fill bag limits for other species.
Beyond the fact that red snapper were a much higher percentage of the catch, even
for 9/0 hooks, than their percent abundance in reef fish community data, another
apparent trend was a decline of smaller intertivore species with increasing hook
size. These smaller species tended to be invertivores, such as red porgy, vermilion
snapper, gray triggerfish, and tomtate (Manooch 1977, Grimes 1979, Thomas and
Cahoon 1993). Some, such as vermilion snapper and red porgy, constituted a much
higher percentage of the 9/0 hook catch than their percent abundance in reef fish
community data. For others, like tomtate, and to a lesser extent, gray snapper, their
Figure 6. Maximum likelihood selectivity functions estimated for red snapper, Lutjanus
campechanus, captured with 9/0, 12/0, and 15/0 circle hooks at reef sites in the northern Gulf of
Mexico.
PATTERSON III ET AL.: CIRCLE HOOK SIZE EFFECT IN REEF FISH FISHERY 659
percent abundance even in 9/0 hook catches was just a fraction of their percent abundance
in the reef fish community. The decline of the catch of smaller fishery species
with increasing hook size, and the percent abundance of the smallest fishery
species, tomtate, being an order of magnitude less abundant in 9/0 catches than in
the community, is likely largely due to gape limitation (Ralston 1982, Bacheler and
Buckel 2004). However, bait selection and fishing technique, which were standardized
and not tested as factors in the present study, also may have affected catch rates.
Cumulative catch rates among scamp, gag, and red groupers, all of which have large
gapes (Weaver 1996), were much greater than their percent abundance in the fish
community, which is consistent with the hypothesis that gape limitation explains
much of the decline in the catch of smaller species with larger hooks.
Figure 7. Predicted vs observed (catch) proportion-at-size of red snapper, Lutjanus campechanus,
captured with 9/0, 12/0, and 15/0 circle hooks at reef sites in the northern Gulf of Mexico.
Combination-1 (9/0 and 12/0), combination-2 (9/0 and 15/0), and combination-3 (12/0 and 15/0)
are abbreviated on panel labels as C1, C2, and C3, respectively. Predicted proportions at size
result from exponential logistic selectivity models fit to observed proportion at size data for each
hook comparison combination.
660 BULLETIN OF MARINE SCIENCE. VOL 88, NO 3. 2012
Potential effects of regulating hook size on red snapper stock recovery, as well
as effects on other co-occurring reef fishes, will remain unclear until results of the
present study and follow-up sampling is incorporated into stock assessment model
simulations. Given the diversity of fishes targeted and caught in the northern GOM
recreational reef fish fishery, and the inverse effect of increasing circle hook size on
catches of species other than red snapper, a multispecies assessment approach likely
would be required, or at least effects should be evaluated through simulations examined
with various single-species assessment models. In the case of red snapper, circle
hook size significantly affected the size of fish captured, with an increasing proportion
of the catch above the minimum size limit as hook size increased, but there was
only a subtle shift in the median size and estimated age of fish captured with 9/0 vs
15/0 hooks. Given the size difference in hooks, a shift in median size from 424 to 482
mm TL and in estimated age from 2.9 to 3.6 yrs is not substantial for a fish that can
live to be 60 yrs old (Patterson et al. 2001, Wilson and Nieland 2001), does not reach
maximum fecundity until it is 12–15 yrs old (Jackson et al. 2007), and for which
maximum yield per recruit occurs at sizes >600 mm TL (Goodyear 1995). However,
the fact that red snapper catch rate did not decline with increasing hook size implies
that individuals were aggressive to the bait regardless of the size of hooks, which
is something that has been reported anecdotally by fishermen attempting to target
other species during red snapper closed seasons. Perhaps gape limitation is less of an
issue for aggressive species such as red snapper.
Another issue to consider when evaluating the potential conservation gains of establishing
a minimum hook size for the GOM recreational reef fish fishery is that
during the sampling for our study, we were informed that some charter and private
boat recreational anglers target reef fishes with circle hooks smaller than the 9/0
hooks used in our study. One of the difficulties in comparing results among hook
studies, or even in understanding fishing practices, is that hook sizes are not standardized
among manufacturers, or even within a given manufacturer for different
hook types. Therefore, future work should focus on examining catch rates, species
composition, and size distributions between hook sizes used in the present study
with some of the smaller hooks currently used in the fishery.
Size distributions estimated in situ with the ROV provided a valuable source of
information for estimating selectivity functions of the 9/0, 12/0, and 15/0 circle
hooks examined. Due to sample sizes, functions were computed only for red snapper,
but the procedure developed for estimating hook selectivity from ROV-based
estimates of fish size distributions on reefs and hook-specific catches could be applied
to other species with sufficient sample sizes. Bacheler et al. (2010) reviewed several
methods of estimating gear selectivity, including internally in stock assessment
models (Hillborn and Walters 1992, Porch 2007), through comparison of catches
between gears (Millar 1992), and with tagging experiments (Schultz 2004, Bacheler
et al. 2008). They concluded that tagging methods were the most robust because the
size or age composition of the tagged population was known, especially in shortterm
tagging experiments. The ROV-based method we employed prior to fishing
similarly provided an estimate of the size distribution of the targeted population.
Furthermore, by controlling hook sizes in fishing trials, we could directly test selectivity
for a given hook size and type, and not just estimate overall selectivity for
fishery sectors (Bacheler et al. 2010).
PATTERSON III ET AL.: CIRCLE HOOK SIZE EFFECT IN REEF FISH FISHERY 661
Selectivity curves for longline hooks and recreational hook-and-line gear have
been assumed to have either logistic shapes typical of trawls or unimodal shapes typical
of gillnets (Czerwinski et al. 2010). The shape of the curve is often imposed with
some prior knowledge of the size distribution of the catch. In the method presented
here, a maximum likelihood framework is used to test the shape of hook-specific selectivity
functions given estimates of the size distribution of the fish being targeted
and direct measurements of the fish caught. A critical assumption of this approach
is that laser-based estimates of fish size are random and unbiased. We applied the
slight bias-correction estimated by Patterson et al. (2009), but the issue of whether
fish scaled with the lasers were a random sample of the population persists. What
is known is that fish tend not to avoid the micro ROVs used in our study (Dance et
al. 2011), and no fish were specifically targeted with the lasers. However, a future
experiment should be conducted in which reef-associated fish communities are repeatedly
sampled on short time scales (e.g., minutes to hours) and estimates of fish
size distributions compared. In the present study, if red snapper smaller than those
reported here were present but not scaled with the ROV lasers, it likely would have
had little effect on selectivity estimates for the hooks tested as the smaller length
bins already had zeros for proportion caught. However, if there were fish present that
were larger than those scaled with lasers, then the shape of the 15/0 hook selectivity
function may have been dome-shaped if the pattern of declining proportional catch
with increasing fish size was maintained.
The difference in shapes among hook-specific selectivity functions estimated here
is due to the fact the β parameter in Equation 4 did not differ significantly from zero
for the 15/0 hook model, thus giving that function a logistic shape. However, when
inspecting predicted vs observed size distributions of 15/0 hook catches, a somewhat
dome-shaped relationship is apparent in the observed catch at size data, especially
for hook combination-2. The divergence from a domed-shape function appears to be
driven by the bimodal pattern of catch-at-size observed for 15/0 hooks fished in combination-
3. There is a sharp decline in the observed catch at size between 420 and 570
mm TL, but then a spike at 610 mm TL. There was an apparent conflict in attempting
to fit that high catch at size for combination-3 data while a different pattern existed
for combination-2. The result was a non-significant β15/0 parameter (i.e., a logistic
shape) and large residuals for the fits of the selectivity function to observed catch at
size for fish >500 mm TL in both combination-2 and combination-3 scenarios. The
logistic fit for 15/0 hooks predicts sizes at median and full selectivity that are similar
to those of 9/0 hooks and smaller than those of 12/0 hooks, which is not consistent
with patterns observed in the data. It should be noted, however, that the 15/0 hooks
had the smallest sample size and perhaps additional field experiments would help
clarify the functional form of their selectivity curve.
The exponential-logistic selectivity model also overestimated the proportion of
small fishes (<400 mm) caught by 9/0 hooks at reef sites where they were fished with
12/0 hooks (combination-1), but underestimated the proportion of small fishes at
sites where 9/0 hooks were fished with 15/0 hooks (combination-2). In principle, the
competing mortality model should account for the fact that the two hook sizes are
competing for fish of the same size. However, it is possible that the behavior of different
fish size classes changes with their relative abundance in a way that affects the
apparent selection. For example, larger, more aggressive fish might drive smaller fish
away or smaller fish might become emboldened at higher densities. The ROV video
662 BULLETIN OF MARINE SCIENCE. VOL 88, NO 3. 2012
data suggest there were proportionally more large fish among reef sites fished with
combination-3 than combination-2, and more at combination-2 sites than combination-
1. In general, a comparison of the residual patterns suggests fewer small fish
were caught on 9/0 hooks than expected at sites with proportionally fewer small fish.
If this trend is real, then it suggests smaller fish actually become more vulnerable
to the gear as they decrease in proportion to larger fish. Additional fishing trials,
especially single hook size instead of hook combination trials, may help clarify this
somewhat counterintuitive result.
In conclusion, results from the present study indicate that varying circle hook size
in the GOM reef fish fishery had significant effects on fish catch rates, species composition,
and size distributions. It is unclear at this point what the conservation benefits
vs detriments would be in requiring a minimum hook size in the fishery, particularly
for red snapper. However, it is clear that increasing hook size does increase the size
of fish captured while also greatly diminishing the diversity of the catch. Future work
should focus on expanding sample sizes in hook trials, increasing the size range of
circle hooks tested, and projecting simulated effects of changing hook sizes and their
associated selectivity functions on biomass trajectories of exploited GOM reef fishes.
Acknowledgments
Funding for this work was provided by the National Marine Fisheries Service’s Cooperative
Research Program (NA09NMF4540137 to WFP). We are grateful to charter boat captains S
Wilson, S Kelley, G Jarvis, and J Greene, and their crews who provided access to their fishing
vessels and considerable help in the field. We also acknowledge the considerable help provided
by J Neese, H Moncrief, R Scharer, M Norberg, J Flynn, and numerous volunteer anglers
for their help in the fishing experiment portion of the study. Lastly, we thank D Nieland and
two anonymous reviewers for helpful comments that improved an earlier version of the paper.
Literature Cited
Alverson DL, Freeberg MH, Pope JG, Murawski SA. 1994. A global assessment of fisheries
bycatch and discards. FAO Fish Tech Paper No. 339. Rome, FAO.
Bacheler NM, Buckel JA. 2004. Does hook type influence the catch rate, size, and injury of
grouper in a North Carolina commercial fishery? Fish Res. 69:303–311. http://dx.doi.
org/10.1016/j.fishres.2004.07.001
Bacheler NM, Hightower JE, Paramore LM, Buckel JA, Pollock KH. 2008. Age-dependent tag
return model to estimate mortality and selectivity of an estuarine-dependent fish with high
rates of catch and release. Trans Am Fish Soc. 137:1422–1432. http://dx.doi.org/10.1577/
T07-146.1
Bacheler NM, Hightower JE, Burdick SM, Paramore LM, Buckel JA, Pollock KH. 2010. Using
generalized linear models to estimate selectivity from short-term recoveries of tagged red
drum Sciaenops ocellatus: effects of gear, fate, and regulation period. Fish Res. 102:266–
275. http://dx.doi.org/10.1016/j.fishres.2009.12.007
Carruthers E, Schneider DC, Neilson JD. 2009. Estimating the odds of survival and identifying
mitigation opportunities for common bycatch in pelagic longline fisheries. Biol Conserv.
142:2620–2630. http://dx.doi.org/10.1016/j.biocon.2009.06.010
Clarke KR, Gorley RN. 2001. PRIMER v5: user manual/tutorial. PRIMER-E, Ltd. Plymouth,
United Kingdom.
Coleman FC, Figueira WF, Ueland JS, Crowder LB. 2004. The impact of United States recreational
fisheries on marine fish populations. Science. 305:1958–1960. PMid:15331771.
http://dx.doi.org/10.1126/science.1100397
PATTERSON III ET AL.: CIRCLE HOOK SIZE EFFECT IN REEF FISH FISHERY 663
Czerwinskia IA, Gutiérrez-Estradab JC, Casimiro-Soriguer-Escofeta M, Hernandoa JA. 2010
Hook selectivity models assessment for black spot seabream: classic and heuristic approaches.
Fish Res. 102:41–49. http://dx.doi.org/10.1016/j.fishres.2009.10.005
Dance MA, Patterson III WF, Addis DT. 2011. Fish community and trophic structure at artificial
reef sites in the northeastern Gulf of Mexico. Bull Mar Sci. 87:301–324. http://dx.doi.
org/10.5343/bms.2010.1040
Davies R, Cripps S, Nickson A. 2009. Defining and estimating global marine fisheries bycatch.
Mar Pol. 33(4):661–672. http://dx.doi.org/10.1016/j.marpol.2009.01.003
Gitschlag GR, Renaud ML. 1994. Field experiments on survival rates of caged and released
red snapper. North Am J Fish Manag. 14(1):131–136. http://dx.doi.
org/10.1577/1548-8675(1994)014<0131:FEOSRO>2.3.CO;2
GMFMC (Gulf of Mexico Fishery Management Council). 2007. Amendment 27 to the reef
fish fishery management plan and amendment 14 to the shrimp management plan. Tampa,
Florida. Available at: http://www.gulfcouncil.org/fishery_management_plans. Accessed
July 2011.
Goodyear CP. 1995. Red snapper in US waters of the Gulf of Mexico. National Marine Fisheries
Service, Southeast Fisheries Science Center, Miami Laboratory, MIA-95/96-05, Miami,
Florida.
Grimes CB. 1979. Diet and feeding ecology of the vermillion snapper Rhomboplites aurorubens
from North Carolina and South Carolina USA waters. Bull Mar Sci. 29:53–61.
Grover AM, Mohr MS, Palmer-Zwahlen ML. 2002. Hook-and-release mortality of Chinook
salmon from drift mooching with circle hooks: management implications for California’s
ocean sport fishery. Am Fish Soc Sym. 30:39–56.
Hanson CW, Sauls B. 2011. Status of recreational saltwater fishing in Florida: characterization
of license sales, participation, and fishing effort. Am Fish Soc Sym. 75:355–365.
Hilborn RA, Walters CJ. 1992. Quantitative fisheries stock assessment: choice, dynamics and
uncertainty. Chapman and Hall, New York. http://dx.doi.org/10.1007/978-1-4615-3598-0
Hurvich CM, Tsai CL. 1995. Model selection for extended quasi-likelihood models in small
samples. Biometrics. 51(3):1077–1084. PMid:7548692. http://dx.doi.org/10.2307/2533006
Ihde TF, Wilberg MJ, Loewensteiner DA, Secor DH, Miller TJ. 2011. The increasing importance
of marine recreational fishing in the US: challenges for management. Fish Res.
108:268–276. http://dx.doi.org/10.1016/j.fishres.2010.12.016
Jackson MW, Cowan Jr JH, Nieland DL. 2007. Demographic differences in northern Gulf
of Mexico red snapper: implications for the unit stock hypothesis. Am Fish Soc Sym.
60:217–228.
Johnson AG, Baker MS, Collins LA. 1997. Magnitude and composition of undersized grouper
bycatch. Proc Gulf Carib Fish Inst. 49:161–172.
Kaplan IC, Cox SP, Kitchell JF. 2007. Circle hooks for Pacific longliners: not a panacea for marlin
and shark bycatch, but part of the solution. Trans Am Fish Soc. 136:392–401. http://
dx.doi.org/10.1577/T05-301.1
Kelleher K. 2005. Discarding in the world’s marine fisheries: an update. Food and Agriculture
Organization of the United Nations. FAO Fish Tech Paper No. 470. Rome.
Lewison RL, Freeman SA, Crowder LB. 2004. Quantifying the effects of fisheries on threatened
species: the impact of pelagic longlines on loggerhead and leatherback sea turtles. Ecol Lett.
7:221–231. http://dx.doi.org/10.1111/j.1461-0248.2004.00573.x
Lindberg WJ, Frazer TK, Portier KM, Vose F, Loftin J, Murie DJ, Mason DM, Nagy B, Hart
MK. 2006. Density-dependent habitat selection and performance by a large mobile reef
fish. Ecol Appl. 16:731–746. http://dx.doi.org/10.1890/1051-0761(2006)016[0731:DHSAP
B]2.0.CO;2
Manooch CS. 1977. Foods of the red porgy, Pagrus pagrus Linnaeus (Pisces: Sparidae), from
North Carolina and South Carolina. Bull Mar Sci. 27:776–787.
Millar RB. 1992. Estimating the size-selectivity of fishing gear by conditioning on the total
catch. J Am Stat Assoc. 87:962–968. http://dx.doi.org/10.2307/2290632
664 BULLETIN OF MARINE SCIENCE. VOL 88, NO 3. 2012
Mitchell KM, Henwood T, Fitzhugh GR, Allman RJ. 2004. Distribution, abundance, and age
structure of red snapper (Lutjanus campechanus) caught on research longlines in US Gulf
of Mexico. Gulf Mex Sci. 22:164–172.
Pacheco JC, Kerstetter DW, Hazin FH, Hazin H, Segundo RSSL, Graves JE, Carvalho F,
Travassos PE. 2011. A comparison of circle hook and J hook performance in a western
equatorial Atlantic Ocean pelagic longline fishery. Fish Res. 107:39–45. http://dx.doi.
org/10.1016/j.fishres.2010.10.003
Patterson III WF. 2007. A review of Gulf of Mexico red snapper movement studies: implications
for population structure. Am Fish Soc Sym. 60:221–236.
Patterson III WF, Dance MA, Addis DT. 2009. Development of a remotely operated vehicle
based methodology to estimate fish community structure at artificial reef sites in the northern
Gulf of Mexico. Proc Gulf Caribb Fish Inst. 61:263–270.
Patterson III WF, Watterson JC, Shipp RL, Cowan Jr JH. 2001. Movement of tagged red
snapper in the northern Gulf of Mexico. Trans Am Fish Soc. 130:533–545. http://dx.doi.
org/10.1577/1548-8659(2001)130<0533:MOTRSI>2.0.CO;2
Porch CE. 2007. An assessment of the red snapper fishery in the US Gulf of Mexico using a
spatially-explicit age-structured model. Am Fish Soc Sym. 60:355–384.
Prince ED, Ortiz M, Venizelos A. 2002. A comparison of circle hook and “J” hook performance
in recreational catch-and-release fisheries for billfish. Am Fish Soc Sym. 30:66–79.
Ralston S. 1982. Influence of hook size in the Hawaiian deep-sea handline fishery. Can J Fish
Aquat Sci. 39:1297–1302. http://dx.doi.org/10.1139/f82-171
Rummer JL. 2007. Factors affecting catch and release (CAR) mortality in fish: insight into CAR
mortality in red snapper and the influence of catastrophic decompression. Am Fish Soc
Sym. 60:113–132.
Sales G, Guffoni B, Fiedler FN, Azevedo VG, Kotas JE, Swimmer Y, Bugoni L. 2010. Circle
hook effectiveness for the mitigation of sea turtle bycatch and capture of target species in
a Brazilian pelagic longline fishery. Aquat Conserv Mar Freshwat Ecosyst. 20(4)428–436.
http://dx.doi.org/10.1002/aqc.1106
SAS Institute, Inc. 1998. Statistics Version 6.11. SAS Institute, Inc., Cary, North Carolina.
Schultz RD. 2004. Size-specific angling return rates of white bass in Kansas reservoirs. N Am J
Fish Manage. 24(2):719–723. http://dx.doi.org/10.1577/M02-100.1
SEDAR (Southeast Data Assessment and Review). 2006. Southeast data, assessment, and review:
stock assessment report of SEDAR 10: Gulf of Mexico gag grouper. Charleston, South
Carolina. Available at: http://www.sefsc.noaa.gov/sedar. Accessed July 2011.
SEDAR (Southeast Data Assessment and Review). 2009a. Stock assessment of red snapper in
the Gulf of Mexico-SEDAR update assessment-Report of the update assessment workshop,
Miami, Florida. Available at: http://www.gulfcouncil.org/about/ftp.php. Accessed July
2011.
SEDAR (Southeast Data Assessment and Review). 2009b. Stock assessment of gag in the Gulf
of Mexico-SEDAR update assessment-Report of the update assessment workshop, Miami,
Florida. Available at: http://www.sefsc.noaa.gov/sedar. Accessed July 2011.
Serafy JE, Cooke SJ, Diaz GA, Graves J, Hall M, Shivji M, Swimmer Y. 2012. Circle hooks in
commercial, recreational, and artisanal fisheries: research status and needs for improved
conservation and management. Bull Mar Sci. 88:371–391. http://dx.doi.org/10.5343/
bms.2012.1038
Strelcheck AJ, Hood PB. 2007. Rebuilding red snapper: recent management activities and future
management challenges. Am Fish Soc Sym. 60:353–363.
Thomas CJ, Cahoon LB. 1993. Stable isotope analyses differentiate between different trophic
pathways supporting rocky-reef fishes. Mar Ecol Prog Ser. 95:19–24. http://dx.doi.
org/10.3354/meps095019
Weaver DC. 1996. Feeding ecology and ecomorphology of three sea basses (Pisces:Serranidae)
in the northeastern Gulf of Mexico. MS thesis, University of Florida. Gainesville, Florida.
PATTERSON III ET AL.: CIRCLE HOOK SIZE EFFECT IN REEF FISH FISHERY 665
Wilson Jr RR, Burns KM. 1996. Potential survival of released groupers caught deeper than
40 m based on shipboard and in situ observations, and tag-recapture data. Bull Mar Sci.
58:234–247.
Wilson CA, Nieland DL. 2001. Age and growth of red snapper, Lutjanus campechanus, from
the northern Gulf of Mexico off Louisiana. Fish Bull US. 99:653–664.
Date SuBmitted: 5 August, 20 11.
Date Accepted: 28 February, 2012.
AVailaBle Online: 25 April, 2012.
Addresses: (WFP, JHT) University of South Alabama and Dauphin Island Sea Lab, 101
Bienville Boulevard, Dauphin Island, Alabama 36528. (CEP) National Marine Fisheries
Service, Southeast Fisheries Science Center, Sustainable Fisheries Division, 75 Virginia Beach
Drive, Miami, Florida 33149. (AJS) NOAA Fisheries Service, Southeast Regional Offi ce, 263
13th Avenue South, St. Petersburg, Florida 33701. Corresponding Author: (WFP) Email:
<wpatterson@disl.org>.