2011 SOUTHEASTERN NATURALIST 10(3):399–408
Growth of the Flattened Musk Turtle, Sternotherus
depressus Tinkle and Webb
Sherry R. Melancon1, Robert A. Angus1, and Ken R. Marion1,*
Abstract - Sternotherus depressus (Flattened Musk Turtle) is a threatened endemic of
the Black Warrior River basin in northern Alabama. Carapace lengths taken from recaptures
of turtles individually marked over an 18-year period were used to generate von
Bertalanffy growth curves for both sexes. Females grew faster than males early in life
and reached estimated asymptotic size at a younger age. Size-at-sexual-maturity data
from previous studies indicate that males reach maturity in 10–12 years, while females
mature at 12–15 years. Compared to a spring-dwelling Florida population of S. minor
(Loggerhead Musk Turtle), the Flattened Musk Turtle has a slower growth rate, a greater
age at maturity, and a smaller asymptotic size. These differing life-history characteristics
likely reflect cooler water temperatures and reduced food availability in Flattened Musk
Turtle habitats.
Introduction
Sternotherus depressus Tinkle and Webb (Flattened Musk Turtle) is a small
aquatic kinosternid turtle endemic to the Black Warrior River drainage above
the Fall Line in northern Alabama (Ernst and Lovich 2009). The turtle prefers
clear, shallow streams with a rocky or sandy substrate. The species is federally
listed as threatened under the Endangered Species Act, primarily because
of strip-mining activities for coal, which have degraded many streams in the
Black Warrior basin (USFWS 1987). According to Dodd (1990), the Flattened
Musk Turtle has disappeared from more than 50% of its former range, and most
remaining populations are fragmented by extensive areas of degraded habitat.
However, some moderate-sized populations still exist in a few locations (Holmes
and Marion 2004). As with many long-lived species, age determination,
life span, growth rates, and sex-specific growth patterns are difficult to ascertain
through direct observations or short-term studies. Knowledge of such
parameters may be important in the conservation of declining or threatened
species or in management for their recovery. In this study, we examine aspects
of the growth of the Flattened Musk Turtle.
Annual rings on epidermal scutes have often been used to estimate the age
of young turtles, but they become indistinct and increasingly unreliable in older
individuals (Germano and Bury 1998). The von Bertalanffy growth equation
has often been used to generate growth curves for several groups of vertebrates
using data from individuals of known age (e.g., Huang et al. 2008). Due to the
inability to obtain reliable age estimates in some organisms, such as older turtles,
1Biology Department, University of Alabama at Birmingham, Birmingham, AL 35294-
1170. *Corresponding author - kmarion@uab.edu.
400 Southeastern Naturalist Vol. 10, No. 3
an alternative method of determining growth patterns was devised. The von
Bertalanffy growth equation was rearranged by Fabens (1965) into a form that
uses growth intervals. The interval form of the equation permits the estimation
of growth parameters using capture-recapture data from animals of unknown
age. Frazer et al. (1990) used a natural population of Trachemys scripta Schoepff
(Pond Slider) to compare a von Bertalanffy growth equation using known age
data to Fabens’ derivative method under the assumption that no ages were known.
The study concluded that there were no significant differences in the values of
any of the estimated parameters.
In order to use Fabens’ derivative method, a measurement at the time of
initial capture and at the time of recapture, and the time interval between
measurements, are required. Fabens’ derivative equation has been used to construct
growth curves for a number of turtle species, including Caretta caretta
L. (Loggerhead Sea Turtle; Casale et al. 2009, Frazer and Ehrhart 1985),
Chelonia mydas L. (Green Sea Turtle; Frazer and Ehrhart 1985, Green 1993),
Emydoidea blandingii Holbrook (Blanding’s Turtle; Huang et al. 2008),
Gopherus polyphemus Daudin (Gopher Tortoise; Aresco and Guyer 1999),
Sternotherus minor Agassiz (Loggerhead Musk Turtle; Cox et al. 1991),
Terrapene carolina L. (Eastern Box Turtle; Dodd and Dreslik 2008), and Trachemys
scripta Schoepff (Pond Slider; Frazer et al. 1990). Although the von
Bertalanffy growth model cannot be used to estimate the age of an individual
turtle with a high degree of precision, it is useful in the absence of more reliable
methods for estimating the approximate age of a turtle of a given sex and
carapace length. Additionally, analysis of von Bertalanffy growth curves can
reveal useful information about the differential life-history strategies which
may be employed by males and females of a species.
An alternative growth model, the Gompertz model, has been used to estimate
growth patterns in some turtles (Cox et al. 1991). Gompertz curves were generated
in the initial phases of this study, but the fits were very poor; therefore, only
the von Bertalanffy model was used in our analysis.
Methods
Study areas and data collection
Carapace length records of individually marked Flattened Musk Turtles from
several mark–recapture studies previously conducted on Sipsey Fork and Brushy
Creek in Winston County, AL were combined with recapture data from the
current study to determine growth intervals and instantaneous individual growth
rates. Sipsey Fork and Brushy Creek are similar-sized confluent tributaries of
the Black Warrior River system. Carapace length records for individual turtles
were available from a 1983 study by Ernst et al. (1989); 1985 and 1986 studies
by Dodd et al. (1988); a report for the Water Works and Sewer Board of the City
of Birmingham, AL by Mount et al. (1991); 1994 and 1995 studies by Bailey
and Guyer (1998); and from the current study conducted in 2002 and 2003. Data
2011 S.R. Melancon, R.A. Angus, and K.R. Marion 401
from each of these studies were compiled, and turtles which had been captured
more than once were identified. Any cases where there was doubt of individual
identification or inconsistency in gender assignment between captures were
discarded. A total of 192 instances of recapture between years were found. In
cases when the same individual was recaptured more than once over the period
of these studies, only the carapace lengths at the first and last captures were used.
The resulting 154 growth intervals (90 males, 61 females, and 3 juveniles) were
used in our analyses.
Turtles in all studies were captured using wire-mesh funnel traps (Iverson
1979a) baited with sardines in soybean oil or were hand collected by wading
at dusk or night with a flashlight. In all studies except the current one, captured
Flattened Musk Turtles were given a unique identification using scute notching;
however, in the current study, an 11-mm passive integrated transponder (PIT) tag
(Biomark, Inc., Boise, ID) was injected into the right anterior inguinal region
parallel to the bridge of the shell to give each captured animal a unique identifi-
cation number (Buhlmann and Tuberville 1998), and any notching from previous
studies was recorded. Carapace length in all studies was measured by calipers to
the nearest 0.1 mm.
Variation in individual growth rates
Growth rates in individual turtles have been shown to be highly variable
(Davenport and Scott 1993, Gibbons et al. 1981, Moll and Legler 1971, Plummer
1977, Sexton 1965). Relative instantaneous growth rates were determined
for each recaptured turtle using the ΔGR equation modified from Brody’s (1945)
derivation using the formula
ΔGR = loge(L2 / L1) / (t2 - t1) = (loge[L2] - loge[L1]) / (t2 - t1),
where L1 and L2 represent the carapace lengths (in mm) at first and second
captures, respectively, and t2 - t1 represents the time interval (in years to
two decimal places) between consecutive measurements. The formula yields
an estimate of instantaneous growth rate at mean carapace length (CL)
that can be used to compare the growth rates of turtles of similar CL’s. The
instantaneous growth rate for each individual was plotted against its mean
size between captures. Linear regression was performed separately for each
sex to investigate trends in growth rates as size increased. The difference
between sexes in the slope of the growth vs. size regression was tested for
significance using a t test (Zar 2010). Outliers with unusually high growth
rates were checked for common features such as area and year of capture. This
procedure was used by Cox et al. (1991) to evaluate variations in growth rates
in Sternotherus minor.
von Bertalanffy growth model
Growth was modeled separately for each sex using the von Bertalanffy
growth equation
402 Southeastern Naturalist Vol. 10, No. 3
L = a(1 - be-kt),
where L is carapace length, a is the asymptotic carapace length, b is a parameter
related to the size at hatching, k is the intrinsic growth rate, and t is age. The a and
k parameters were estimated using the von Bertalanffy growth interval equation
for capture-recapture data
L2
= a - (a - L1)e-kd,
where L1 is CL at first capture, L2 is the CL at recapture, and d is the time between
captures (Aresco and Guyer 1991, Cox et al. 1991, Frazer and Ehrhart 1985).
The r2 (observed vs. predicted) value was used to indicate the goodness-of-fit
to the data. After the asymptotic size and intrinsic growth rate parameters were
estimated, b was estimated using the formula
b = 1 - (h / a),
where h is the carapace length at hatch (28 mm; based on the average of two
clutches measured two days after hatching). Ninety-five percent confidence
intervals were constructed around the growth curves using the method of Seber
and Wild (1989).
To test the hypothesis that the growth parameters were different for males and
females, a dummy variable (z) was coded to be 0 for females and 1 for males.
The growth interval equation was fitted to the total data set (males and females)
using nonlinear least squares regression, but with an extra two “difference”
parameters (da and dk) that allowed the asymptote and intrinsic growth rate to
take on different values in males and females:
L2
= (a + da x z) - ([a + da x z] - L1) x e-(k + dk x z) x d.
If the 95% confidence interval for the estimate of da or dk did not contain zero,
the corresponding parameter (a or k) was considered to differ significantly between
sexes at α = 0.05.
Due to the primarily insectivorous diet of smaller juvenile Flattened Musk
Turtles (Tinkle 1958), only a few juveniles were captured by baited traps during
all of the various studies. Thus, the generated growth curves might not
accurately estimate the rate of growth during the first few years of life. In order
to validate the generated growth curves and the 95% confidence intervals,
eight preserved smaller juveniles (ranging from 34–55 mm CL) were obtained
from collections at the University of Alabama and Auburn University. The age
of each individual was estimated by counting growth annuli on the carapacial
scutes using the technique of Sexton (1959). These individuals of undetermined
sex were then plotted on the growth curve for each sex. The results of
this study were compared to those of a similar study on the closely-related
Sternotherus minor (Cox et al. 1991). The S. minor study was conducted in a
warm-water spring in Florida with an abundant food supply, whereas the Flattened
Musk Turtle occurs in cooler-water streams with abundant shade, where
activity is suspended during the winter.
2011 S.R. Melancon, R.A. Angus, and K.R. Marion 403
Results
Instantaneous growth rate
The amount of time between first capture and last capture for the 154 individuals
used in the analysis averaged 3.8 years and ranged from 1 to 18 years. ΔGR
was negatively correlated with mean size between the initial capture and recapture
in both sexes (females: r = -0.60, P = 3.63 × 10-7; males r = -0.26, P = 0.014;
Fig. 1). The decrease in growth rate with size was steepest in females. However,
due to the large amount of variability in the data, the slopes of the regression lines
did not differ significantly between the sexes (P = 0.106). Capture location and
year of capture were not predictors of the outliers.
von Bertalanffy growth model
Figure 2 shows the von Bertalanffy growth curves for both sexes. The r2
(observed vs. predicted) for males was 0.96 and for females was 0.95, both
indicating good fits. The estimated asymptotic CLs (a) of male and female
Flattened Musk Turtles were 102.7 and 102.1 mm, respectively. The Wald
95% confidence interval for the difference variable, da, included zero, indicating
that the difference in asymptotic size between males and females was not
significant at α = 0.05; however, males and females do not reach their maximum
sizes at the same rate. The intrinsic growth rate parameter (k) estimate
for males was 0.04 and for females 0.07. The Wald 95% confidence interval
Figure 1. Relative instantaneous growth rates of individual turtles (ΔGR) plotted against
mean carapace length (CL) between initial capture and recapture.
404 Southeastern Naturalist Vol. 10, No. 3
Figure 2. von Bertalanffy growth curves for male (a) and female (b) Flattened Musk
Turtles. CL is carapace length (mm). Dashed lines represent 95% confidence limits.
Darkened circles represent juveniles from museum specimens of undetermined sex
whose ages have been estimated from growth annuli on carapacial scutes. Open circles
represent two superimposed individuals.
2011 S.R. Melancon, R.A. Angus, and K.R. Marion 405
for the difference variable, dk, did not contain zero, indicating that the difference
was significant at α = 0.05.
Length and age estimates from museum specimens of smaller unsexed juveniles
are indicated as dots on the growth curves of both sexes (Fig. 2). These
individuals fall within the 95% confidence intervals for females. Two of the eight
fall outside the confidence intervals for male growth, perhaps indicating that the
curve may slightly underestimate early growth rate in males or that those individuals
were females.
Discussion
Differential growth rates between the sexes have been noted in several species
of turtles (Ernst and Lovich 2009, Gibbons et al. 1981). Our results indicate that
female Flattened Musk Turtles grow faster early in life than do males, but reach
the same asymptotic adult size. Female Flattened Musk Turtles lay 1–3 large eggs
per clutch up to two times per year (Close 1982). The eggs can be larger than 30
mm x 15 mm and can weigh more than 5 grams (Dodd et al. 1988). The accelerated
growth rate of females early in life may represent an adaptive strategy to
reach egg-bearing size as soon as possible, thereby minimizing the risk of juvenile
female mortality. In contrast, the generated growth model indicates that male
growth throughout life is more gradual. Male Flattened Musk Turtles have larger
home ranges than females (Dodd et al. 1988). In order to maximize their reproductive
potential, males may spend more energy searching for potential mates
and less energy on growth. Consequently, females reach 95% of their asymptotic
size at 38.0 years of age, while males do not reach 95% of their asymptotic size
until 61.0 years of age (Fig. 2).
Based on histological examination of gonads, Close (1982) conservatively
estimated that male Flattened Musk Turtles reach maturity around 60 mm CL,
while females become sexually mature between 70 mm and 75 mm CL. Based
on the generated growth curves (Fig. 2), males should reach sexual maturity at
10–12 years of age, while females become sexually mature in 12–15 years. These
estimates of time to reach maturity are somewhat greater than for most other
kinosternid turtles (Ernst and Lovich 2009; Iverson 1979b, 1991), and may reflect
the relatively unproductive environmental conditions of their habitat. Cox et al.
(1991) applied the von Bertalanffy and Gompertz growth models to a springdwelling
Florida population of the closely related Loggerhead Musk Turtle, and
estimated that females reached maturity in 8 years at a mean of 80 mm CL, while
males averaged 5.6 years to reach maturity at 55 mm CL. Additionally, both sexes
essentially reached their asymptotic size between 25 and 30 years of age, earlier
than for Flattened Musk Turtles. Loggerhead Musk Turtles also had a higher instantaneous
growth rate and reached a greater asymptotic size (male = 112.8 mm
CL, female = 110.4 mm CL). Such differences in growth characteristics between
Flattened Musk Turtles and the population of Loggerhead Musk Turtles studied
by Cox et al. (1991) likely represent physiological and evolutionary responses
to differing environmental conditions. The Loggerhead Musk Turtle study was
406 Southeastern Naturalist Vol. 10, No. 3
conducted in a relatively warm-water spring, where the turtles have a long activity
period and an abundant mollusk food supply. Our study was conducted in a
temperate stream with abundant shade cover where activity is suspended during
the winter. In addition, mollusk populations in Sipsey Fork and Brushy Creek are
low to moderate in density (Rogers and Marion 2004).
The Flattened Musk Turtle has suffered significant population declines, local
extirpation, and habitat fragmentation over most of its geographic range,
primarily due to habitat degradation caused by previous coal strip-mining activities
(Dodd 1990). Fonnesbeck and Dodd (2003) also found that a disease
of unknown etiology reduced population size on a stretch of the Sipsey Fork
in the mid-1980s. They noted that population levels had not recovered significantly
over time. Indeed, life-history characteristics exhibited by Flattened
Musk Turtles indicate that conservation and management goals for the species
(i.e., population increases and/or repopulation in recovering streams) will require
an extended period of time.
Acknowledgments
This study was part of the senior author’s Master’s Thesis research at the University
of Alabama at Birmingham. We sincerely thank C. Kenneth Dodd, Jr. and Karan
Bailey and Craig Guyer for providing us with meristic measurement information on
turtles marked in their previous studies. We thank Craig Guyer and Reid Downer
for providing access to preserved specimens at the Auburn University Natural History
Museum and the University of Alabama Scientific Collections, respectively.
The US Forest Service provided partial financial support. Tom Counts and Allison
Baker, US Forest Service, Bankhead National Forest, provided significant logistical
support for our field studies.
Literature Cited
Aresco, J.M., and C. Guyer. 1999. Growth of the tortoise Gopherus polyphemus in slash
pine plantations of south central Alabama. Herpetologica 55:499–506.
Bailey, K.A., and C. Guyer. 1998. Demography and status of the Flattened Musk Turtle,
Sternotherus depressus, in the Black Warrior River basin of Alabama. Chelonian
Conservation and Biology 3:77–83.
Brody, S. 1945. Bioenergetics and Growth. Reinhold Publications Corp., New York, NY.
Buhlmann, K.A., and T. Tuberville. 1998. Use of passive integrated transponder (PIT)
tags for marking small freshwater turtles. Chelonian Conservation and Biology
3:102–104.
Casale, P., A.D. Mazaris, D. Freggi, C. Vallini, and R. Argano. 2009. Growth rates and
age at adult size of Loggerhead Sea Turtles (Caretta caretta) in the Mediterranean Sea,
estimated through capture-mark-recapture records. Scientia Marina 73(3):589–595.
Close, D.K. 1982. The reproductive cycle of Sternotherus minor depressus. Unpubl.
M.Sc. Thesis. University of Alabama at Birmingham, AL.
Cox, W.A., J.B. Hazelrig, M.E. Turner, R.A. Angus, and K.R. Marion. 1991. A model
for growth in the musk turtle, Sternotherus minor, in a north Florida Spring. Copeia
1991:954–968
2011 S.R. Melancon, R.A. Angus, and K.R. Marion 407
Davenport, J., and C.R. Scott. 1993. Individual growth and allometry of young Green
Turtles (Chelonia mydas L). Herpetological Journal 3:19–25
Dodd, C.K., Jr. 1990. Effects of habitat fragmentation on a stream-dwelling species, the
Flattened Musk Turtle, Sternotherus depressus. Biological Conservation 54:33–45.
Dodd, C.K., Jr., and M. Dreslik. 2008. Habitat disturbances differentially affect individual
growth rates in a long-lived turtle. Journal of Zoology 275(1):18–25.
Dodd, C.K., Jr., K. Enge, and J. Stuart. 1988. Aspects of the biology of the Flattened
Musk Turtle, Sternotherus depressus, in northern Alabama. Bulletin of the Florida
State Museum, Biological Sciences 34:1–64.
Ernst, C.H., and J.E. Lovich. 2009. Turtles of the United States and Canada. Second Edition.
The Johns Hopkins University Press, Baltimore, MD. Pp. 514–518.
Ernst, C.H., W. Cox, and K. Marion. 1989. The distribution and status of the Flattened
Musk Turtle, Sternotherus depressus (Testudines: Kinosternidae). Tulane Studies in
Zoology and Botany 27:1–20.
Fabens, A.J. 1965. Properties and fitting of the von Bertalanffy growth curve. Growth
29:265–289.
Fonnesbeck, C.J., and C.K. Dodd, Jr. 2003. Estimation of Flattened Musk Turtle (Sternotherus
depressus) survival, recapture, and recovery rate during and after a disease
outbreak. Journal of Herpetology 37:602–607.
Frazer, N.B., and L.M. Ehrhart. 1985. Preliminary growth model for Green, Chelonia
mydas, and Loggerhead, Caretta caretta, Turtles in the wild. Copeia 1985:73–79.
Frazer, N.B., J.W. Gibbons, and J.L. Greene. 1990. Exploring Fabens’ growth interval
model with data on a long-lived vertebrate, Trachemys scripta. Copeia 1990:112–118.
Gibbons, J.W., R.D. Semlitsch, J.L. Greene, and J.P. Schubauer. 1981. Variation in age
and size at maturity of the Slider Turtle (Pseudemys scripta). American Naturalist
117:841–845.
Germano, D.J., and R.B. Bury. 1998. Age determination in turtles: Evidence of annual
deposition of scute rings. Chelonian Conservation and Biology 3:123–132.
Green, D. 1993. Growth rates of wild immature Green Turtles in the Galapagos Islands,
Ecuador. Journal of Herpetology 27:338–341.
Holmes, S., and K. Marion. 2004. The status of the populations of the Flattened Musk
Turtle (Sternotherus depressus) in Bankhead National Forest and Smith Lake, Alabama,
and the possible effects of stream conditions on trapping success. Southeastern
Biology 51:143–144.
Huang, Y.C., Y. Zhang, P. Cabilo, M. Richard, and T. Herman. 2008. Analysis of the
growth of the Nova Scotia Blanding’s Turtle. Atlantic Electronic Journal of Mathematics
3(1):18–29.
Iverson, J.B. 1979a. Another inexpensive turtle trap. Herpetological Review 10:55.
Iverson, J.B. 1979b. Reproduction and growth of the Mud Turtle, Kinosternon subrubrum
(Reptilia, Testudines, Kinosternidae), in Arkansas. Journal of Herpetology
13:105–111.
Iverson, J.B. 1991. Life history and demography of the Yellow Mud Turtle, Kinosternon
flavescens. Herpetologica 47:373–395.
Moll, E.O., and J.M. Legler. 1971. The life history of a neotropical slider turtle, Pseudemys
scripta (Schoepff), in Panama. Bulletin of the Los Angeles County Museum of
Natural History, Science 11:1–102.
Mount, R.H., K.R. Marion, and W.A. Cox. 1991. Status of the Flattened Musk Turtle,
Sternotherus depressus, in the mid-reaches of the Locust Fork of the Black Warrior
River, Blount and Jefferson counties, Alabama. Report to the Water Works and Sewer
Board of the City of Birmingham, AL. 49 pp.
408 Southeastern Naturalist Vol. 10, No. 3
Plummer, M.V. 1977. Reproduction and growth in the turtle Trionyx muticus. Copeia
1977:440–447.
Rogers, S.R.H, and K.R. Marion. 2004. Assessment of the suitability of selected streams
in Bankhead National Forest for occupation by populations of Flattened Musk Turtles
(Sternotherus depressus), and the potential effects of the silvicultural improvements
on habitat quality. Report to the USDA Forest Service, Double Springs, AL; Alabama
Power Company, Environmental Services, Birmingham, AL; and the Nature Conservancy
of Alabama, Birmingham, AL. 170 pp.
Seber, G.A.F., and C.J. Wild. 1989. Nonlinear Regression. John Wiley and Sons, Inc.,
New York, NY.
Sexton, O.J. 1959. A method for estimating the age of painted turtles for use in demographic
studies. Ecology 40:716–718.
Sexton, O.J. 1965. The annual cycle of growth and shedding in the Midland Painted
Turtle, Chrysemys picta marginata. Copeia 1965:314–318.
Tinkle, D.W. 1958. The systematic and ecology of the Sternotherus carinatus complex
(Testudinata, Chelydridae). Tulane Studies in Zoology and Botany 6:1–56.
United States Fish and Wildlife Service (USFWS). 1987. Determination of threatened
status for the Flattened Musk Turtle (Sternotherus despressus). Federal Register 52
(112):22418–22430.
Zar, J.H. 2010. Biostatistical Analysis. Fifth Edition. Prentice Hall Publishers, Upper
Saddle River, NJ.