Content uploaded by Will Selman
Author content
All content in this area was uploaded by Will Selman on Jan 17, 2017
Content may be subject to copyright.
BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors, nonprofit publishers, academic institutions, research
libraries, and research funders in the common goal of maximizing access to critical research.
Population Structure, Status, and Conservation of Two Graptemys Species from the
Pearl River, Mississippi
Author(s): Will Selman and Robert L. Jones
Source: Journal of Herpetology, 51(1):27-36.
Published By: The Society for the Study of Amphibians and Reptiles
DOI: http://dx.doi.org/10.1670/15-082
URL: http://www.bioone.org/doi/full/10.1670/15-082
BioOne (www.bioone.org) is a nonprofit, online aggregation of core research in the biological, ecological, and
environmental sciences. BioOne provides a sustainable online platform for over 170 journals and books published
by nonprofit societies, associations, museums, institutions, and presses.
Your use of this PDF, the BioOne Web site, and all posted and associated content indicates your acceptance of
BioOne’s Terms of Use, available at www.bioone.org/page/terms_of_use.
Usage of BioOne content is strictly limited to personal, educational, and non-commercial use. Commercial inquiries
or rights and permissions requests should be directed to the individual publisher as copyright holder.
Journal of Herpetology, Vol. 51, No. 1, 27–36, 2017
Copyright 2017 Society for the Study of Amphibians and Reptiles
Population Structure, Status, and Conservation of Two Graptemys Species from the
Pearl River, Mississippi
WILL SELMAN
1,2,3
AND ROBERT L. JONES
4
1
Department of Biological Sciences, University of Southern Mississippi, 118 College Drive, 5018, Hattiesburg, Mississippi, USA
2
Department of Biology, Millsaps College, 1701 North State Street, Jackson, Mississippi, USA
4
Mississippi Department of Wildlife, Fisheries, and Parks, Museum of Natural Science, 2148 Riverside Drive, Jackson, Mississippi, USA
ABSTRACT.—Graptemys is one of the least studied turtle genera in North America. Graptemys oculifera (Ringed Sawback) and
Graptemys pearlensis (Pearl Map Turtle) are endemic to the Pearl River system of Mississippi and southeastern Louisiana. We studied
both species near Columbia, Mississippi, on the Pearl River via a trapping and basking surveys over two years. Additionally, five sites
including Columbia were trapped for 27 years to determine long-term trends in capture success and relative abundance (RA). At the
Columbia site, body size distribution was bimodal for G. oculifera and atypically unimodal for G. pearlensis;G. pearlensis body lengths
were smaller than museum specimens. Population estimates for G. oculifera at the Columbia site indicate a stable population over 25
years. Long-term RA trends indicated that G. pearlensis was less common than G. oculifera in all periods and at all sites from 1988 to 2013.
Trends in long-term capture success for G. oculifera and G. pearlensis were negative at all sites, with significant declines at three sites for
both G. oculifera and G. pearlensis. Declines occurred both upstream and downstream of a major reservoir. Therefore, a combination of
factors (including altered hydrology, human disturbance, lack of recruitment, excessive sedimentation, impaired water quality, and/or the
pet trade) appear to be contributing to declines. Additional conservation and protection is warranted for G. pearlensis, and current
protections for G. oculifera should be extended. Future studies should continue at our long-term sites to determine whether population
declines persist or whether populations stabilize.
Understanding wildlife population sizes and demography are
central tenets for species conservation efforts; however, these
data are lacking for many southeastern freshwater turtle species
to aid in conservation efforts. In particular, Graptemys (map
turtles and sawbacks) is considered one of the least studied
turtle genera in North America (Lovich and Ennen, 2013); yet
many species within the genus are of conservation concern
(Buhlmann and Gibbons, 1997; van Dijk et al., 2014) and lack
basic life-history information (Lindeman, 2013). Further, most of
the species solely inhabit riverine environments and associated
water bodies (e.g., oxbows and bayous), with nine of the 14
species endemic to single river drainages of the Gulf of Mexico
(Lindeman, 2013). Many of the endemic Gulf Coast Graptemys
are considered the least understood within the group (Lovich
and Ennen, 2013).
Two such species, Graptemys oculifera (Ringed Sawback; Baur,
1890) and Graptemys pearlensis (Pearl Map Turtle; Ennen et al.,
2010), occur sympatrically in the Pearl River system of central
Mississippi and southeastern Louisiana. Considerable knowl-
edge is available for G. oculifera including population densities
(Jones and Hartfield, 1995; Dickerson and Reine, 1996; Linde-
man, 1998; Shively, 1999), population structure (Jones and
Hartfield, 1995), reproductive ecology (Jones, 2006), and
population genetics (Gaillard et al., 2015). Most of this
information was made possible because the species was listed
as federally threatened in 1986 (USFWS, 1986) and with the
undertaking of proposed studies outlined in the G. oculifera
recovery plan (Stewart, 1988). Most of the available data for G.
pearlensis, however, is coincidental to visual population density
surveys focused on G. oculifera (Dickerson and Reine, 1996;
Lindeman, 1998; Shively, 1999), with all data reported prior to
its recognition as a separate Graptemys taxon (Ennen et al., 2010).
Presumably, the lack of data available for G. pearlensis is because
1) it previously had a much wider range when considered as
either G. pulchra (sensu lato; Lovich and McCoy, 1992) or G.
gibbonsi (sensu lato; Ennen et al., 2010); 2) it was perceived to
need less conservation attention, because it was not a drainage
endemic as either G. pulchra (sensu lato) or G. gibbonsi (sensu
lato); and 3) it was overshadowed by the federally threatened
status of G. oculifera. The latter directed more funding resources
toward G. oculifera research to understand basic life-history and
ecological characteristics, with data for G. pearlensis only
collected secondarily.
The first objective in this study was to document in-depth
population structure and densities of both G. oculifera and G.
pearlensis at one site on the lower Pearl River of southern
Mississippi. The second objective was to examine trends in
trapping success (i.e., catch per unit effort, CPUE) and relative
abundance (RA) at five sites throughout the Pearl River system
in Mississippi for 27 yr. The third goal was to compare the data
collected in this study to historical data for both species within
the drainage.
MATERIALS AND METHODS
Study Sites.—For the detailed demographic study, we conduct-
ed fieldwork on 8.3 river kilometers (rkm) of the Pearl River near
the town of Columbia (Marion County, MS; Fig. 1). The
Columbia site is characteristic of a medium-sized (~75–175 m
wide), Gulf Coastal Plain river, with alternating pointbar and
cutbank sections, abundant submergent and emergent dead-
wood snags, and a sand and gravel substrate. The flow of the
Pearl River (discharge: 24–1700 m
3
/sec) is primarily regulated by
the Ross Barnett Reservoir spillway, ~240 rkm upstream from the
site. The site is surrounded by riparian bottomland forest (i.e.,
Water Oak [Quercus nigra], Baldcypress [Taxodium distichum],
Sycamore [Platanus occidentalis], Spruce Pine [Pinus glabra]), as
well as a small amount of pasture land and a small number of
fishing camps. The eastern bank at the southern end of the site is
bordered by a local park, which is a county-maintained and -
3
Corresponding author. E-mail: will.selman@millsaps.edu
DOI: 10.1670/15-082
operated park with a boat launch, ball fields, and pavilions.
Boating levels on the river are relatively low even though the
boat ramp provides river access.
For the long-term study (1988 and 2014), we trapped over
multiple sampling periods, not including the 2005/2006
trapping at Columbia described below. This includes at the
Columbia site (N=6 sampling periods) and four other sites—
Carthage (6), Ratliff Ferry (7), Lakeland (7), and Monticello (7)—
to determine long-term trends in species CPUE and RA (Fig. 1).
Two sites are located upstream of the Ross Barnett Reservoir
(Carthage, Ratliff Ferry), and three sites are located downstream
(Lakeland, Monticello, Columbia). These study sites are
described in more detail by Jones and Hartfield (1995).
Short-Term Study.—During 2005–2006, Graptemys species were
trapped at the Columbia site for 2–4 days per month from April
to October. We trapped turtles by attaching open topped basking
traps (made of 3/4-inch PVC coated crawfish wire; Fish Net
Company, Jonesville, LA) to emergent deadwood and left them
slightly submerged (Selman et al., 2012). Traps varied in size
(from 56 ·46 ·31 cm to 122 ·61 ·25 cm) and were attached
with nails and cotton twine to known Graptemys basking logs or
branches used by both sexes and different size classes. Trap sites
were later approached rapidly via motorized boat, which startled
FIG. 1. The geographic location of the Pearl River in the southeastern United States (top inset) and map of sample sites in central Mississippi
(bottom). Cross-hatching represents areas where Graptemys oculifera and Graptemys pearlensis co-occur, whereas stippling represents upstream areas
only occupied by G. pearlensis (based on maps by Lindeman, 2013) and new records of Lindeman (2014a, b).
28 W. SELMAN AND R. L. JONES
basking turtles into the traps. A maximum of 17 traps were used
during a trap-day, and each trap was checked approximately
every hour; traps were occasionally adjusted because of
fluctuating water levels or moved if turtles avoided the trap site.
Turtles also were captured opportunistically by hand or by dip
net.
After capture, male G. oculifera turtles were determined by
smaller body sizes and longer foreclaws, taller carapacial spines,
and longer tails (Jones and Selman, 2009). Graptemys pearlensis
males were smaller, did not have a greatly enlarged head, and
had longer tails compared to females (Lovich et al., 2009).
Midline plastron length (PL) was measured to the nearest
millimeter with tree calipers, and body mass (in grams) was
measured with a hanging scale (Ohaus, Parsippany, NJ). We
palpated females to determine their reproductive status and
permanently marked turtles with holes drilled into marginal
scutes (Cagle, 1939). During October in 2005 and 2006, we
marked G. oculifera on the carapace (second and third vertebral
scutes) with a waterproof, tree-marking spray paint (Aervoe
t
Lead-Free Fluorescent Glo Spray Paint, Gardnerville, NV) for
subsequent mark–resight surveys.
We also conducted basking density surveys at the Columbia
site for G. oculifera in October 2005 and for both Graptemys
species in October 2006. The site was surveyed by walking
sandbars from the downstream end to the upstream end using a
60 mm, 15–45X spotting scope with tripod to locate basking
turtles. Population densities were calculated by counting the
number of turtles observed and dividing it by the total length of
river surveyed (2.75 rkm); the midriver distance was measured
using the GoogleEarth path measuring tool (v. 6.1; Google, Inc.,
Mountain View, CA). In 2006, RA was calculated for only
Graptemys species, but other basking species are reported.
During all surveys, paint-marked G. oculifera were resighted,
and all surveys were completed within two weeks of the initial
paint mark application to assure that no paint marks were lost
(Selman and Qualls, 2008; Kornilev et al., 2012).
Long-Term Study.—Along with the detailed data collected at the
Columbia site during 2005 and 2006, Graptemys were trapped at
five sites between 1988 and 2014 to document long-term CPUE
and RA trends using similar trapping methods described above.
Each day, we used 28 to 30 traps throughout the entire study and
used a mean of 29 traps/day for calculations. We tallied the
number of captured G. oculifera and G. pearlensis, along with
morphological data for G. oculifera as described elsewhere (Jones
and Hartfield, 1995; Jones, 2006). These data permitted us to
calculate yearly and daily CPUE and RA for each species at each
site. The former method permits comparisons of densities over
time, whereas the latter method has been used in other studies as
an indicator of species rank abundance relative to other
sympatric species (Lindeman, 1998, 1999, 2013).
Statistical Analyses.—For the short-term study at the Columbia
site, we used t-tests for both species to compare male and female
plastron lengths. To determine the degree of sexual size
dimorphism for each species, we calculated the sexual dimor-
phism index (SDI; Lovich and Gibbons, 1992):
SDI=Mature female PL
Male PL
-1:
We use a Chi-square contingency table to determine whether
populations deviated from a 1 : 1 sex ratio. Population size
estimation via mark–resighting of paint-marked G. oculifera was
completed in 2005 and 2006. Similar to methods used by Selman
and Qualls (2008, 2009), we used program NOREMARK (White,
1996), because it did not require uniquely marked turtles, and
additional marked individuals could be added to the popula-
tion between survey intervals.
For the long-term study, we used a linear regression to
analyze the RA of G. pearlensis for all sites over the 27-yr time
period; G. oculifera RA trend would be the inverse relationship.
We also used linear correlations to analyze daily capture success
over the 27-yr period for both species at all sites. We used JMP
9.0 for all statistical analyses (SAS Institute, Inc., Cary, NC,
http://www.jmp.com) except for the linear correlations for
daily capture success where we used SYSTAT (Systat Software,
Inc., Chicago, IL; http://www.Systat.com).
RESULTS
Short-Term Study.—Turtles were captured on 33 trapping days
at the Columbia site across 13 months (April through October) in
2005 and 2006. This excludes September 2005 immediately
following Hurricane Katrina. During this period, 198 G. oculifera
(107 M, 91 F) and 39 G. pearlensis (28 M, 10 F, 1 juvenile) were
captured.
Graptemys oculifera exhibited a typical bimodal size class
distribution (Fig. 2), with females attaining significantly greater
size than that of males (t
197
=-34.88, P<0.0001) and strong
female-biased sexual-size dimorphism (SDI =0.67; Table 1). The
sex ratio of G. oculifera did not deviate from equality (v
2
=1.13,
df =1, P=0.29). The mean size for gravid G. oculifera females
was 12.9 cm PL (N=10; range: 11.6–14.0 cm PL), with all gravid
females captured between early May (10 May 2006) and mid-
July (14 July 2005).
Graptemys pearlensis exhibited an atypical unimodal size class
distribution, primarily associated with a small number of female
captures (N=10) across a wide range of size classes (PL range:
11.2–18.2; Fig. 2). Female G. pearlensis attained significantly
greater size than males (t
36
=-5.63, P=0.0003; Table 1) and
also exhibited strong female-biased sexual size dimorphism
(SDI =0.88). The sex ratio of G. pearlensis was significantly
skewed toward males (v
2
=8.8, df =1, P=0.0029). No gravid
G. pearlensis females were captured, with only 3 moderately
sized, nongravid females (13.7, 17.0, 18.2 cm PL) captured
between May and July for both years.
For October basking surveys, mean basking densities of G.
oculifera in 2006 were nearly double those observed in 2005
(2005 mean: 36.5 per rkm, 2006 mean: 62.1 per rkm; Table 2).
Even though basking densities showed high variability across
the two survey years, population estimates via mark–resight
methods were nearly identical (Table 2). Graptemys oculifera RA
in 2006 was high (mean: 0.79, range: 0.79–0.80), and they were
the dominant basking turtle species at the site.
Population estimates for G. pearlensis via mark–resight could
not be completed because we were unable to capture and paint-
mark enough individuals in a short period of time to meet the
assumption of a closed population; however, G. pearlensis mean
basking density in 2006 was 16.1 per rkm (range: 14.9–18.1 per
rkm), and G. pearlensis RA (mean: 0.21, range: 0.20–0.22) was
much lower than that of G. oculifera. Other turtle species
observed in 2006 surveys include Pseudemys concinna (River
Cooter; mean density =10.3 per rkm), Trachemys scripta (Slider
Turtle; 0.6 per rkm), Sternotherus carinatus (Razorback Musk
Turtle; 0.6 per rkm), and Apalone sp. (softshell turtle species; 0.2
per rkm).
Long-Term Study.—Between 1988 and 2013, RA was much
greater for G. oculifera than for G. pearlensis at all five sites and
PEARL RIVER GRAPTEMYS STATUS AND CONSERVATION 29
across all years (Table 3, Fig. 3). The northernmost site near
Carthage had declining CPUE trends for both Graptemys species,
but the G. oculifera trend of decline was significant (Table 3).
Carthage, however, had the most equitable ratios between the
two species, but G. oculifera RA (0.69–0.83) was always greater
than that of G. pearlensis RA (0.17–0.31). The next site
downstream, Ratliff Ferry, had significant declining CPUE trends
for both Graptemys species (Table 3). Ratliff Ferry also had the
most skewed species ratios with G. oculifera (0.95–1.0) vastly
outnumbering G. pearlensis (0.00–0.05) in all years sampled. For
the middle site, Lakeland, both species had declining CPUE
trends, but the G. pearlensis trend of decline was significant (Table
3). At Lakeland, G. oculifera RA (0.67–0.97) was always greater
than G. pearlensis (0.03–0.33), with G. pearlensis nearly absent from
the site by the end of the study period (from 0.33 to 0.03). For the
next two sites downstream, Monticello and Columbia, declining
trends in CPUE were all negative, but it was a significant decline
for both species only at the Columbia site (Table 3). For these two
sites, RA was highly skewed toward G. oculifera (Monticello:
0.78–1.0; Columbia: 0.67–0.73) relative to G. pearlensis (Monticello:
0.22–0.0; Columbia: 0.27–0.33). With all sites considered through-
out the study, the RA trend for G. pearlensis was negative but not
significant (F
1,36
=0.52, P=0.47).
Comparisons of the two long-term data sets indicate that RA
and CPUE did not always align (Fig. 3A, 3B). Both data sets
indicated declining trends for G. pearlensis at Ratliff Ferry,
Lakeland, and Columbia, but RA and CPUE data did not align
at Carthage (CPUE–ns decline; RA–increase) or Monticello
(CPUE–ns decline; RA–stable). Both RA and CPUE data
indicated declines for G. oculifera at Carthage but did not align
at Ratliff Ferry (CPUE–sig decline; RA–increase), Lakeland
(CPUE–ns decline; RA–increase), Monticello (CPUE–ns decline;
RA–stable), or Columbia (CPUE–sig decline; RA–increase).
DISCUSSION
Short-Term Study.—When comparing our G. oculifera data to
Jones and Hartfield (1990, 1995) at the same Columbia site using
the same basking trap methods, mean body lengths for both
males and all females are slightly greater, but the SDI for this
study (0.67) is similar to that of 1990 (0.70; Jones and Hartfield
1995). We captured no unsexable juveniles in 198 captures (0%) in
2005/2006, compared to 15 juveniles in 362 captures in 1989/
1990 (4%; Jones and Hartfield 1995). This appears to be an aging
population attributable to limited recent recruitment as evi-
denced by larger adult body sizes and a lack of juveniles
compared to 25 yr prior.
Very few G. pearlensis were captured to accurately document
population structure at the Columbia site, and no gravid
females were sampled, although no other prior data are
available to compare either of these metrics. Male G. pearlensis
mean (7.4 cm PL) and maximum body lengths (9.2 cm PL) were
small compared to larger males observed in other localities
(pers. obs.) and in museum collections (Ennen et al., 2010; P.
Lindeman, pers. comm.). Even more noticeable was the low
numbers of female captures (N=10 over 2 yrs) with no
definitive size class distribution. Similar to males, female G.
pearlensis mean (14.1 cm PL) and maximum body lengths (18.2
cm PL) were small compared to females from other locations
throughout the Pearl River system (pers. obs.) and museum
specimens (P. V. Lindeman, pers. comm.; Table 1). For the latter,
Ennen et al. (2010) noted the maximum size for a G. pearlensis
female was 29.5 cm CL, which equates to ~25 cm PL (based on a
mean 0.85 PL:CL ratio for the 10 females in this study); this
would be approximately 40% greater than the largest individual
encountered at the Columbia site (18.2 cm PL). Particularly
lacking at the Columbia site were the reproductive size classes
(N=3; 16.9 cm, 17.5 cm, 18.2 cm) as defined by Cagle (1952;
>17 cm max PL, ~16 cm midline PL), which are also
represented in museum specimen collections (P. V. Lindeman,
pers. comm.). The three basking surveys in 2006 similarly
indicated a male-skewed population with only 15 females
observed compared to 123 males.
Using identical mark–resight survey methodology at the
same site, Jones and Hartfield (1995) found that G. oculifera
population estimates (131 per rkm) were similar to this study
(146–151 per rkm). There are not similar mark–resight data for
G. pearlensis, but our densities for both G. pearlensis (16.1 per
rkm) and G. oculifera (62.1 per rkm) were threefold higher than
those described by Lindeman (1998, 1999) for the region around
FIG. 2. Number of captures for different size classes of Graptemys
oculifera (top) and Graptemys pearlensis (bottom) from the Columbia site
(Marion County, MS). Unknown sex juveniles are in black, males in light
gray, and females in dark gray. The black arrows indicate the size classes
of gravid females in G. oculifera found at the site and the smallest gravid
female G. pearlensis found by Cagle (1952).
30 W. SELMAN AND R. L. JONES
the Columbia site (5.1 G. pearlensis per rkm, 20.4 G. oculifera per
rkm). Even though absolute densities differed, RA values from
Lindeman’s study (1 G. pearlensis: 4.0 G. oculifera) are similar to
our observations (1 G. pearlensis: 3.9 G. oculifera). The discrep-
ancy between these studies’ density values is likely because
surveys occurred during different seasons (fall for this study,
summer for Lindeman’s studies) and the river distance sampled
(boat surveys, walking, and counting sandbars for this study,
fixed point counts for Lindeman’s studies). Thus, higher
basking densities in this study were expected, because of
decreased fall air and water temperatures that promoted higher
basking frequencies compared to the summer (Selman and
Qualls, 2011). Using basking density surveys, Dickerson and
Reine (1996) also found G. oculifera to be more abundant than G.
pearlensis at all nine sites on the Pearl River, with G. oculifera
densities ranging from 10 to 22 times higher than densities of G.
pearlensis. Shively (1999) encountered fewer G. pearlensis (N=
370, 22% of basking turtles) than G. oculifera (N=513, 30%) on
Bogue Chitto River in Louisiana, a tributary of the Pearl River.
Therefore, all investigators over the last 20 yr have found G.
pearlensis to be a smaller component of the Pearl River basking
turtle community. For future studies, mark–resight population
estimates would be preferable, because of the considerable
variation observed in basking densities across time of day,
month, and environmental conditions (Jones and Hartfield,
1995; Selman and Qualls, 2011).
Long-Term Study.—Our long-term trapping RA data indicated
that G. oculifera was more abundant than G. pearlensis at all sites
and during all sampling periods. However, because using RA
comparisons over long periods has limitations (i.e., assumptions
about similar population growth rates and direct competition),
inferring actual population density changes using this data set is
TABLE 1. Comparisons of male and female plastron lengths (cm) and body mass (g) for Graptemys oculifera and Graptemys pearlensis from the
Columbia site (Marion County, MS). Body-size comparisons for G. pearlensis are for museum specimens throughout the species’ range.
G. oculifera G. pearlensis
This study Jones and Hartfield (1990, 1995) This study P. V. Lindeman, pers. comm.
Male Female Male Female Male Female Male Female
PL
N107 92 138 115 28 10 70 21
Mean 7.8 12.6 7.4 10.6 9.3 14.1 8.1 16.0
Min 5.6 8.0 6.1 6.1 8.1 11.2 5.8 7.1
Max 9.1 15.4 9.2 14.5 10.2 18.2 9.8 21.5
SE 0.06 0.13 – – 0.10 0.85 0.11 1.09
Mass
N107 92 28 10
Mean 96.7 437.4 178.6 740.5
Min 50 110 105 290
Max 140 775 235 1500
SE 1.72 13.22 6.79 147.1
TABLE 2. Population density and relative abundance (RA) of Graptemys species from the short-term study site at Columbia site (Marion County,
MS) and historical studies (organized by year). Basking density and mark–resight values are expressed as number of turtles per river kilometer.
Method of sampling is next to the study name; methods include basking density surveys (BD) without marked individuals and mark–resight surveys
(MR) for Graptemys oculifera with paint-marked individuals. For the latter, confidence levels are included in parenthesis below the population estimate.
Study Site
Survey month
and year
Graptemys oculifera Graptemys pearlensis
Number
observed
Basking
density RA
Mark–resight
population estimate
Number
observed
Basking
density RA
Jones and Hartfield (1990) MR Columbia July 1995 131 (84–211)
Lindeman Pearl River May and June 23.6 0.91 2.4 0.09
(1998) BD S. Pearl River 1994, 1995 20.4 0.80 5.1 0.20
Dickerson Ratliff Ferry July 1996 924 83.2 0.92 75 6.8 0.08
and Reine Lakeland ‘‘ 389 52.6 0.96 18 2.4 0.04
(1996) BD Pools Bluff ‘‘ 11 1.4 0.92 1 0.1 0.08
Pools Bluff Sill ‘‘ 10 3.0 0.91 1 0.3 0.09
Bogue Chitto Sill ‘‘ 58 15.7 0.94 4 1.1 0.06
Lock no. 1 ‘‘ 101 13.7 0.94 7 1.0 0.06
I-59 ‘‘ 43 2.9 1.00 0 0 0
I-10 ‘‘ 3 1.6 1.00 0 0 0
Hwy 90 ‘‘ 0 0 0 0 0 0
Shively (1999) BD Bogue Chitto May–July 1999 513 4–17 0.58 370 2–15 0.42
This Study Columbia October 2005 96 34.9
BD, MR ‘‘ 106 38.5
‘‘ 99 36.0
2005 Mean 100 36.5 146 (94–259)
October 2006 181 65.8 0.78 50 18.1 0.22
‘‘ 166 60.4 0.80 42 15.3 0.20
‘‘ 165 60.0 0.80 41 14.9 0.20
2006 Mean 171 62.1 0.80 151 (107–230) 44 16.1 0.20
PEARL RIVER GRAPTEMYS STATUS AND CONSERVATION 31
difficult. Therefore, we also used CPUE data as a comparison to
determine 1) long-term trends and 2) whether differences occur
between the two methods. Even though RA comparisons showed
that G. oculifera appeared to be doing ‘‘relatively well’’ compared
to G. pearlensis, our CPUE data indicated that both species were
in decline at all sites over the 27-yr period, with six of the 10
species by site comparisons indicating a significant decline. For
the latter, three of these were for G. oculifera (Carthage, Ratliff
Ferry, Columbia), and three were for G. pearlensis (Ratliff Ferry,
Lakeland, Columbia). Interestingly, the CPUE data indicated that
declines of both species have occurred both upstream and
downstream of the Ross Barnett Reservoir (RBR). There are many
potential explanations for the observed demographic irregular-
ities we observed for the G. pearlensis population and for declines
of both species downstream of the RBR. These include 1)
impaired water quality attributable to industrial and/or munic-
ipal effluents, 2) associated impacts of reservoir flow regulation,
3) collection by the pet trade, or 4) a combination of these.
First, industrial, agricultural, and municipal effluents may
have historically impaired water quality in the lower Pearl
River. Effluents have degraded water quality and negatively
impacted freshwater mussels and gastropods (Neves et al.,
1997), likely primary food items for large male and female G.
pearlensis (Lovich et al., 2009; pers. obs.) but not major foods
items for G. oculifera (Kofron, 1991). In the late 1970s, McCoy
and Vogt (1979) noted severe sewage pollution in the Pearl
River downstream of Jackson, with no Graptemys observed
between the towns of Jackson and Terry (~43 rkm). Presumably,
water quality in the Pearl River has improved since 1975 with
wastewater and sewer infrastructure improvements in the
vicinity of Jackson (Mississippi Department of Environmental
Quality, 1998). McCoy and Vogt (1979) also noted heavy
discharges from paper mills in Monticello, with the water being
‘‘brown and foamy’’ in this area. They concluded that the
cumulative effects of these sources may cause chemical
accumulation in Graptemys, leading to death or impaired
reproduction while also indirectly impacting invertebrate prey
densities; however, the irregularities in the population demo-
graphics may be attributable and a relic of these municipal or
industrial impacts. Shelby and Mendon ¸ca (2001) reported a
severe pulp mill discharge that led to direct mortality of turtles
(unspecified species) in the Leaf and Pascagoula Rivers
(Mississippi), with a similar spill causing mortality in G.
oculifera from the lower Pearl River (Finn, 2011).
Second, the RBR, (constructed in 1963) may have had a
negative impact on downstream Graptemys populations due to
habitat alterations. The RBR maintains a relatively stable water
level throughout the year (within ~0.6 m; USGS Water
Resources, Station 02485600) such that, during lower flow
periods (e.g., typically during the summer months), very little
water is released to downstream reaches of the Pearl River
(Station 02485601). In this and other ways, reservoirs alter
riverine hydrology (for review, see Bunn and Arthington, 2002)
and are a leading contributor to species endangerment in the
United States, particularly in the southeastern United States
(Czech et al., 2000). More specifically, freshwater mollusks, a
primary food item of G. pearlensis (Lindeman, 2013), may be
impacted below reservoirs attributable to changes in flow
regimes (Neves et al., 1997). The Mobile River system of
Alabama, Mississippi, and Georgia is one of the most
impounded river systems in the United States, with high
imperilment of freshwater mollusks (Neves et al., 1997;
Williams et al., 2008). Throughout the Mobile River drainage,
TABLE 3. Number of trap days (TD), captures (Go or Gp), and annual species capture success (Go SorGp S) of Graptemys oculifera (Go) and Graptemys pearlensis (Gp) by sample site within the Pearl
River system of Mississippi. Sites are ordered from upstream (left) to downstream (right). The asterisks (*) indicate significant correlations (corr.) of daily capture success over time.
Year
Carthage Ratliff Ferry Lakeland Monticello Columbia
TD Go Go S Gp Gp STD Go Go SGp Gp STD Go Go SGp Gp STD Go Go SGp Gp STD Go Go SGp Gp S
1988 232 449 1.94 5 0.02 377 246 0.65 41 0.11 377 355 0.94 57 0.15
1989 319 158 0.50 33 0.10 116 271 2.34 8 0.07 116 79 0.68 37 0.32 116 64 0.55 18 0.16 348 282 0.81 133 0.38
1990 116 69 0.59 21 0.18 203 293 1.44 15 0.07 203 78 0.38 21 0.10 174 152 0.87 39 0.22 174 114 0.66 57 0.33
1994 87 70 0.80 15 0.17 87 126 1.45 4 0.05 87 46 0.53 4 0.46 87 78 0.90 21 0.24 87 68 0.78 29 0.33
2002 87 32 0.37 9 0.10 87 113 1.30 3 0.03 87 50 0.58 8 0.09 87 67 0.77 5 0.06 87 41 0.47 15 0.17
2008 87 78 0.89 2 0.02 87 75 0.86 5 0.06 87 40 0.46 2 0.02
2009 87 34 0.39 15 0.17 87 27 0.31 10 0.12
2013 116 89 0.77 0 0.0 116 31 0.27 1 0.01 116 58 0.50 13 0.11
2014 116 22 0.19 7 0.06 116 49 0.42 19 0.16
Total 812 385 100 928 1419 37 1,073 607 117 1,044 814 155 899 581 263
Mean 0.47 0.13 1.45 0.04 0.56 0.10 0.71 0.14 0.58 0.25
Corr. -0.566 -0.185 -0.361 -0.346 -0.034 -0.366 -0.295 -0.204 -0.389 -0.363
t-value 3.36 0.921 2.15 2.05 0.20 2.29 1.78 1.20 1.293 2.025
P0.003* 0.37 0.04* 0.05* 0.84 0.03* 0.09 0.24 0.04* 0.05*
32 W. SELMAN AND R. L. JONES
G. pulchra (Alabama Map Turtle), a similar mollusk specialist, is
observed at much lower RAs (0.15) when compared to the more
generalist G. nigrinoda (Black-Knobbed Sawback, 0.85; Godwin,
2003). Furthermore, when water is released from the RBR, it is
often done rapidly; once water levels approach the target
reservoir level, the flow is quickly stopped. This type of dam
operation contributes to waterlogged banks downstream that
collapse under their own weight when water recedes too
rapidly. Substantial channel filling follows, and we have
observed this at downstream sites over the 27-yr study.
Third, G. pearlensis (particularly females) may have been
captured for the pet trade in the lower Pearl River, whereas
collectors may have been deterred from capturing G. oculifera,
because of their state (Mississippi Department of Wildlife,
Fisheries, and Parks, 2000) and federal (USFWS 1986) listing
status. Because G. pearlensis is one of the few Graptemys species
not protected by state or federal listing, many turtles of all age
classes are offered for sale at pet expositions (C. Lechowitz,
pers. comm.) and online classified pages (pers. obs.); however,
very little information exists on the trade of G. pearlensis.A
FIG. 3. Catch per unit effort (CPUE; panel A) and relative abundance (RA; panel B) of Graptemys oculifera (dashed lines) and Graptemys pearlensis
(solid lines) from five sites sampled over 27 yr in Mississippi. Each site is represented by a different color with site colors the same for both panels.
PEARL RIVER GRAPTEMYS STATUS AND CONSERVATION 33
source known to the authors stated in 2007 that almost all of
the G. pearlensis currently on the market were collected from
the Pearl River by a single collector. Therefore, our numbers
and lack of females in the Columbia population may reflect the
impacts of over-collecting, especially if females were more
highly sought by collectors. This site may have been impacted
by the presence of a nearby public boat ramp. Similarly,
Conway-Gomez (2007) found that river turtle populations
closer to human settlements in the Amazon were more likely to
be exploited than those farther from settlement.
Along with the previously described scenarios downstream
of the RBR, declines also were observed upstream of the RBR.
Declines of both species during the study period at Ratliff Ferry
may be attributable to increased human recreational boating on
the river and extended human presence on nesting sandbars
(i.e., semipermanent camps during the summer months). For the
former, Selman et al. (2013) found that G. flavimaculata, a similar
species in the Pascagoula River, was impacted by boating
recreation through behavioral changes (e.g., limited basking)
and physiological changes (e.g., increased long-term stress).
Additionally, it seems likely that more direct mortalities may be
occurring in this population attributable to faster and larger
boats using the river than in the 1980s and 1990s (RLJ, pers.
obs.). For the impacts of human presence on sandbars, Moore
and Seigel (2006) found limited opportunities for nesting in G.
flavimaculata attributable to long-term human disturbance on
nesting beaches.
Upstream at Carthage, declines of G. oculifera may be
attributable to excessive sedimentation. Tuscolameta Creek
empties into the Pearl River in the lower section of the site
and was channelized in the late 1950s to facilitate drainage
along ~40 km of the creek. The creek has been classified as an
impaired stream attributable to organic enrichment and oxygen
depletion, nutrient load, and sedimentation (Mississippi De-
partment of Environmental Quality, 2014). Channelization has
increased sediment input into the Pearl River since the
beginning of the study and subsequently decreased the
availability of deeper water habitats preferred by Graptemys
(RLJ, pers. obs.).
Historical Comparisons.—Comparisons of our data to historical
data from the Pearl River are difficult to make because of
differing sampling methodologies. Historical efforts used night
hand/dipnet capture methods between the 1950s and 1970s
(Cagle, 1953; Tinkle, 1958; Cliburn, 1971), and these studies found
G. pearlensis were about twice as abundant as G. oculifera.This
method, however, may not be representative of the population,
with the specimen record indicating hand/dipnet sampling was
biased toward juveniles (WS, pers. obs.). Hence, these sampling
efforts may have other biases and misrepresent the true historical
densities of the two Graptemys species. Later collections using
fyke nets indicated a more equitable capture record between the
two Pearl River Graptemys species (105 G. oculifera:80G.
pearlensis; Vogt, 1980). This sample is an intermediate point in
time between earlier hand capturing samples and our later
basking trap samples. Our data from 1988 to 2014 at all sites and
all time periods indicated that G. oculifera RA was always higher
than G. pearlensis, and our CPUE data indicate G. pearlensis
declines over the last 27 yr at three of the five study sites.
Therefore, G. pearlensis may have declined between the 1950s and
mid-1980s, and declines may continue at some sites to the
present.
Graptemys pearlensis also may have been a naturally rare
member of the Pearl River community compared to G. oculifera.
In other Gulf of Mexico river drainages with two Graptemys
species, broad-headed species are outnumbered by the sympat-
ric narrow-headed species, except for headwater reaches of river
systems (the Mobile Basin, Godwin, 2003; Pascagoula River
system, Selman and Qualls, 2009; Selman and Lindeman, 2015;
Sabine and Calcasieu River systems, Louque, 2014). Because
broad-headed species are generally scarcer than sympatric
narrow-headed species, environmental and demographic sto-
chastic events (Lande, 1993) will have greater impacts on them
compared to the more abundant narrow-headed species.
Ultimately, chances of population extinction are higher in
species with smaller populations like G. pearlensis and broad-
headed Graptemys species.
Conservation Implications.—Based on identical sampling meth-
odology as Jones and Hartfield (1995), G. oculifera appear to have
maintained a stable population over 25 yr at the Columbia site,
but currently it appears to be an aging population. The
population structure of G. pearlensis depicts a struggling
population with few reproductively mature females. Numerous
factors may be responsible for these irregularities in the
population, and these factors could have been historical (e.g.,
municipal effluents and water quality) or continue to the present
(e.g., impacts of reservoir on hydrology). Our long-term RA data
indicate that G. oculifera has outnumbered G. pearlensis at all sites
over the last 27 yr, whereas our long-term CPUE data indicate
that populations of both species have declined at all sites, and
significantly at some sites. Data sets for RA and CPUE align only
40% of the time; hence, we suggest caution when using only RA
data for species comparisons attributable to the difficulty in
inferring actual population densities.
Based on our data and comparisons to historical data, G.
pearlensis population declines may have occurred prior to our
study, and declines at some sites have continued to the present;
likewise, G. pearlensis also may represent a naturally less
abundant member of the Pearl River turtle fauna. Regardless
of the exact scenario, all evidence indicates that additional
state/federal protections, comprehensive surveys, and studies
are warranted for G. pearlensis. First, rangewide surveys for G.
pearlensis should focus on documenting their current distribu-
tion and abundance, with particular attention to smaller rivers
and creeks within the system that may harbor populations. New
localities and drainages may be found similar to G. gibbonsi in
the Pascagoula River system (Selman and Qualls, 2009) and
other Graptemys species (Lindeman, 2013). Second, extensive
field efforts should be initiated for G. pearlensis that simulta-
neously document 1) population size via mark–resight efforts,
2) population structure, 3) diet, and 4) aspects of reproduction.
Third, G. oculifera long-term declines at all sites (three being
significant) indicate that continued protection and monitoring
for this species remains warranted. Fourth, because turtles have
long generation times and population trends may not be
noticeable without longer-term data (>30 yr), additional
trapping efforts using similar methods at our study sites should
determine if both G. oculifera and G. pearlensis populations
continue declining or eventually stabilize. These studies would
make great strides in closing the existing life history and
ecology gaps for both species, while also providing valuable
information for future conservation of Graptemys and river
management efforts.
Acknowledgments.—We thank the Mississippi Department of
Wildlife, Fisheries, and Parks (MDWFP) for providing project
funding. This project could not have been completed without
34 W. SELMAN AND R. L. JONES
the technical assistance of T. Mann (Mississippi Museum of
Natural Science), L. McCoy (MDWFP), and C. Qualls (Univer-
sity of Southern Mississippi [USM]). R. Bufkin of Bufkin Marine
(Lucedale, MS) also kept the boat running throughout the 2005
and 2006 field seasons and helped us on many occasions to get
back out on the river. P. Lindeman also provided helpful
information on G. pearlensis museum specimens and measure-
ment data. R. Elsey and two anonymous reviewers provided
helpful comments on an earlier draft of the manuscript. This
project was approved by the USFWS, MDWFP, and the USM
Institutional Animal Care and Use Committee (IACUC
07032201).
LITERATURE CITED
BAUR, G. 1890. Two new species of tortoises from the south. Science
(New York) 16:262–263.
BUHLMANN, K. A., AND J. W. GIBBONS. 1997. Imperiled aquatic reptiles of
the southeastern United States: historical review and current
conservation status. Pp. 201–232 in G. Benz and D. E. Collins
(eds.), Aquatic Fauna in Peril: The Southeastern Perspective. Lenz
Design and Communications, USA.
BUNN, S. E., AND A. H. ARTHINGTON. 2002. Basic principles of ecological
consequences of altered flow regimes for aquatic biodiversity.
Environmental Management 30:492–507.
CAGLE, F. R. 1939. A system of marking turtles for future identification.
Copeia 1939:170–173.
———. 1952. The status of the turtles Graptemys pulchra Baur and
Graptemys barbouri Carr and Marchand, with notes on their natural
history. Copeia 1952:223–234.
———. 1953. The status of the turtle Graptemys oculifera (Baur).
Zoologica 38:137–144.
CLIBURN, J. W. 1971. The ranges of four species of Graptemys in
Mississippi. Journal of Mississippi Academy of Sciences 16:16–19.
CONWAY-GOME
´Z, K. 2007. Effects of human settlements on abundance of
Podocnemis unifilis and P. expansa turtles in northeastern Bolivia.
Chelonian Conservation and Biology 6:199–205.
CZECH, B., P. R. KRAUSMAN,AND P. K. DEVERS. 2000. Economic associations
among causes of species endangerment in the United States.
BioScience 50:593–601.
DICKERSON, D. D., AND K. J. REINE. 1996. Habitat assessment and relative
abundance estimates for the ringed sawback turtle (Graptemys
oculifera) in dredging sites of the West Pearl River navigation project
(Louisiana). Report to the U.S. Army Corps of Engineers. Vicksburg,
MS.
ENNEN, J. R., J. E. LOVICH,B.R.KREISER,W.SELMAN,AND C. P. QUALLS. 2010.
Genetic and morphological variation between populations of the
Pascagoula map turtle (Graptemys gibbonsi)inthePearland
Pascagoula Rivers with description of a new species. Chelonian
Conservation and Biology 9:98–113.
FINN, K. 2011. Louisiana paper mill spill causes massive fish kill.
Available at http://www.reuters.com/article/2011/08/22/
us-louisiana-fishkill-idUSTRE77L6BL20110822. Archived by Web-
Cite at http://www.webcitation.org/6a5yY0osW.
GAILLARD, D. L., W. SELMAN,R.L.JONES,B.R.KREISER,C.P.QUALLS,AND K.
LANDRY. 2015. High connectivity observed in populations of ringed
sawbacks, Graptemys oculifera, in the Pearl and Bogue Chitto Rivers
using six microsatellite loci. Copeia 103:1075–1085.
GODWIN, J. C. 2003. Alabama map turtle (Graptemys pulchra) status
survey. Report to the Alabama Department of Conservation and
Natural Resources. Montgomery.
JONES, R. L. 2006. Reproduction and nesting of the endangered ringed
map turtle, Graptemys oculifera, in Mississippi. Chelonian Conserva-
tion and Biology 5:195–209.
JONES, R. L., AND P. D. HARTFIELD. 1990. Density and population structure
of the ringed sawback turtle, Graptemys oculifera (Baur): year two.
Report to the Mississippi Department of Wildlife, Fisheries, and
Parks, Jackson.
———. 1995. Population size and growth in the turtle Graptemys
oculifera. Journal of Herpetology 29:426–436.
JONES, R. L., AND W. SELMAN. 2009. Graptemys oculifera Baur 1890–ringed
map turtle, ringed sawback. Pp. 033.1–033.8 in A. G. J. Rhodin, P. C.
H. Pritchard, P. P van Dijk, R. A. Saumure, K. A. Buhlmann, J. B.
Iverson, R. A. and Mittermeier (eds.), Conservation Biology of
Freshwater Turtles and Tortoises: A Compilation Project of the
IUCN/SSC Tortoise and Freshwater Turtle Specialist Group.
Chelonian Research Monographs No. 5, doi:10.3854/crm.5.033/
oculifera.v1.2009.
KOFRON C. P. 1991. Aspects of the ecology of the threatened ringed
sawback turtle, Graptemys oculifera. Amphibia–Reptilia 12:161–168.
KORNILEV, Y. I., C. K. DODD JR., AND G. R. JOHNSTON. 2012. Retention of
paint markings for individual identification of free-ranging basking
aquatic turtles (Suwannee cooters, Pseudemys concinna suwanniensis).
Herpetological Review 43:61–64.
LANDE, R. 1993. Risks of population extinction from demographic and
environmental stochasticity and random catastrophes. American
Naturalist 142:911–927.
LINDEMAN, P. V. 1998. Of deadwood and map turtles (Graptemys): an
analysis of species status for five species in three river drainages
using replicated spotting-scope counts of basking turtles. Chelonian
Conservation and Biology 3:13–141.
———. 1999. Surveys of basking map turtles Graptemys spp. in three
river drainages and the importance of deadwood abundance.
Biological Conservation 88:33–42.
———. 2013. The Map Turtle and Sawback Atlas: Ecology, Evolution,
Distribution, and Conservation. University of Oklahoma Press, USA.
———. 2014a. Geographic distribution: Graptemys oculifera. Herpetolog-
ical Review 45:280.
———. 2014b. Geographic distribution: Graptemys pearlensis. Herpeto-
logical Review 45:280.
LOUQUE, I. J., JR. 2014. Status and morphological variation of the Sabine
map turtle (Graptemys sabinensis). Unpubl. master’s thesis, McNeese
State University, USA.
LOVICH, J. E., AND J. R. ENNEN. 2013. A quantitative analysis of the state of
knowledge of turtles of the United States and Canada. Amphibia–
Reptilia 34:11–23.
LOVICH, J. E., AND J. W. GIBBONS. 1992. A review of techniques for
quantifying sexual size dimorphism. Growth, Development, and
Aging 56:269–281.
LOVICH, J. E., AND C. J. MCCOY. 1992. Review of the Graptemys pulchra
group (Reptilia: Testudines: Emydidae), with descriptions of two
new species. Annals of the Carnegie Museum 61:293–315.
LOVICH, J. E., W. SELMAN,AND C. J. MCCOY. 2009. Graptemys gibbonsi
Lovich and McCoy 1992—Pascagoula map turtle, Pearl River map
turtle, Gibbons’ map turtle. Pp. 029.1–029.8 in A. G. J. Rhodin, P. C.
H. Pritchard, P. P van Dijk, R. A. Saumure, K. A. Buhlmann, J. B.
Iverson, and R. A. Mittermeier (eds.), Conservation Biology of
Freshwater Turtles and Tortoises: A Compilation Project of the
IUCN/SSC Tortoise and Freshwater Turtle Specialist Group.
Chelonian Research Monographs No. 5, doi:10.3854/crm.5.029.
gibbonsi.v1.2009.
MCCOY, C. J., AND R. C. VOGT. 1979. Distribution and population status of
the ringed sawback (Graptemys oculifera), blotched sawback (Grap-
temys flavimaculata) and black-knobbed sawback (Graptemys nigrino-
da) in Alabama and Mississippi. Final Report, U.S. Fish and Wildlife
Contract No. 14-16-0004-79-038.
MISSISSIPPI DEPARTMENT OF ENVIRONMENTAL QUALITY.1998.Stateof
Mississippi water quality assessment. Jackson.
———. 2014. State of Mississippi water quality assessment 2014 Section
305(b) Report. Jackson.
MISSISSIPPI DEPARTMENT OF WILDLIFE,FISHERIES,AND PARKS. 2000. Endan-
gered Species of Mississippi. Public Notice No. 3357.001. Mississippi
Department of Wildlife, Fisheries, and Parks, Jackson.
MOORE, M. J. C., AND R. A. SEIGEL. 2006. No place to nest or bask: effects
of human disturbance on the nesting and basking habits of yellow-
blotched map turtles (Graptemys flavimaculata). Biological Conserva-
tion 130:386–393.
NEVES, R. J., A. E. BOGAN,J.D.WILLIAMS,S.A.AHLSTEDT,AND P. W.
HARTFIELD. 1997. Status of aquatic mollusks in the southeastern
United States: a downward spiral of diversity. Pp. 43–86 in G. Benz
and D. E. Collins (eds.), Aquatic Fauna in Peril: the Southeastern
Perspective. Lenz Design and Communications, USA.
SELMAN,W.,AND P. V. LINDEMAN. 2015. Life history and ecology of the
Pascagoula map turtle (Graptemys gibbonsi). Herpetological Conser-
vation and Biology 10:791–800.
SELMAN,W.,AND C. QUALLS. 2008. Impacts of Hurricane Katrina on a
population of yellow-blotched sawbacks (Graptemys flavimaculata)in
the Lower Pascagoula River. Herpetological Conservation and
Biology 3:224–230.
PEARL RIVER GRAPTEMYS STATUS AND CONSERVATION 35
———. 2009. Distribution and abundance of two imperiled Graptemys
species of the Pascagoula River system. Herpetological Conservation
and Biology 4:171–184.
———. 2011. Basking ecology of the yellow-blotched sawback (Grap-
temys flavimaculata), an imperiled turtle species of the Pascagoula
River system, Mississippi, USA. Chelonian Conservation and
Biology 10:188–197.
SELMAN, W., J. M. JAWOR ,AND C. P. QUALLS. 2012. Seasonal variation of
corticosterone levels in Graptemys flavimaculata,animperiled
freshwater turtle. Copeia 2012:698–705.
SELMAN, W., C. QUALLS,AND J. OWEN. 2013. Effects of human disturbance
on the behavior and physiology of an imperiled freshwater turtle.
Journal of Wildlife Management 77:877–885.
SHELBY,J.A.AND M. T. MENDON ¸CA. 2001. Comparison of reproductive
parameters in male yellow-blotched map turtles (Graptemys flavima-
culata) from a historically contaminated site and a reference site.
Comparative Biochemistry and Physiology Part C 129:233–242.
SHIVELY, S. H. 1999. 1999 survey for the ringed map turtle (Graptemys
oculifera) in the Bogue Chitto River, Louisiana. Report to the
Louisiana Department of Wildlife and Fisheries, Baton Rouge.
STEWART, J. H. 1988. A Recovery Plan for the Ringed Sawback Turtle
Graptemys oculifera. U.S. Fish and Wildlife Service, USA.
TINKLE, D. W. 1958. The systematics and ecology of the Sternotherus
carinatus complex (Testudinata, Chelydridae). Tulane Studies in
Zoology 6:3–56.
USFWS (U.S. FISH A ND WILDLIFE SERVICE). 1986. Determination of
threatened status for the ringed sawback turtle. Federal Register
51:45907–45910.
VAN DIJK, P. P., J. B. IVERSON ,A.G.J.RHODIN,H.B.SHAFFER,AND R. BOUR.
2014. Turtles of the world. 7th ed. Annotated checklist of taxonomy,
synonymy, distribution with maps, and conservation status. Pp. 329–
479 in A. G. J. Rhodin, P. C. H. Pritchard, P. P van Dijk, R. A.
Saumure, K. A. Buhlmann, J. B. Iverson, and R. A. Mittermeier (eds.),
Conservation Biology of Freshwater Turtles and Tortoises: A
Compilation Project of the IUCN/SSC Tortoise and Freshwater
Turtle Specialist Group. Chelonian Research Monographs No. 5,
doi:10.3854/crm.5.000.checklist.v7.2014.
VOGT, R. C. 1980. New methods for trapping aquatic turtles. Copeia
1980:368–371.
WHITE, G. C. 1996. NOREMARK: Population estimation from mark–
resighting surveys. Wildlife Society Bulletin 24:50–52.
WILLIAMS, J. D., A. E. BOGAN,AND J. T. GARDNER. 2008. Freshwater Mussels
of Alabama and the Mobile Basin in Georgia, Mississippi, and
Alabama. University of Alabama Press, USA.
Accepted: 1 September 2016.
Published online: 11 January 2017.
36 W. SELMAN AND R. L. JONES
