Atrazine Exposure Impacts Behavior and Survivorship of
Neonatal Turtles
Author(s): Lorin A. Neuman-Lee and Fredric J. Janzen
Source: Herpetologica, 67(1):23-31. 2011.
Published By: The Herpetologists' League
DOI: 10.1655/HERPETOLOGICA-D-09-0003.1
URL: http://www.bioone.org/doi/full/10.1655/HERPETOLOGICA-D-09-0003.1
BioOne (www.bioone.org) is an electronic aggregator of bioscience research content,
and the online home to over 160 journals and books published by not-for-profit 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.
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.
Herpetologica, 67(1), 2011, 23–31
E 2011 by The Herpetologists’ League, Inc.
ATRAZINE EXPOSURE IMPACTS BEHAVIOR AND SURVIVORSHIP OF
NEONATAL TURTLES
LORIN A. NEUMAN-LEE1,2,3
1
AND
FREDRIC J. JANZEN1
Department of Ecology, Evolution and Organismal Biology, Iowa State University, Ames, IA 50011, USA
ABSTRACT: Atrazine (2-chloro-4-ethythlamino-6-isopropylamine-1,3,5-tiazine) is a widely used preemergent herbicide for controlling broadleaf plants. Because atrazine (a known endocrine-disrupting chemical) is
applied in the late spring and early summer, its incidental effects on species exposed to runoff from terrestrial
sources in this time period are of special interest. To examine the possible secondary impact of atrazine, we
obtained eggs from 10 nests of two map turtle species, Graptemys ouachitensis and G. pseudogeographica,
that nest on riverine sandbars. We incubated two eggs from each nest in sand containing one of four initial
concentrations of atrazine (control and 0.1, 10, and 100 mg/L) based on levels measured in the river at the site
where eggs were collected. We recorded hatching success, incubation time, external morphological
abnormalities, gonadal sex, three measures of body size, righting time, and swimming time for all turtles. We
reared a subset of the original neonates individually for 11 mo, during which time nest escape behavior, time
to first foraging event, time to capture prey, growth, and survival were evaluated. None of the variables
recorded at hatching was significantly affected by atrazine treatment, although abnormalities declined as
atrazine levels increased. However, turtles deriving from the lowest atrazine-treated eggs had inhibited nest
escape behavior and reduced posthatching survival. These findings reveal persistent fitness-reducing impacts
on neonatal turtles resulting from atrazine exposure during embryonic development.
Key words: Atrazine; Endocrine disruptor; Graptemys; Reptiles; Turtles
McCoy, 2010; Solomon et al., 2008). This is
surprising given that many reptile species are
highly aquatic, use habitats near agricultural
areas, are carnivores or scavengers, and are
long-lived (e.g., Hopkins, 2000; Moll and
Moll, 2004; Saumure and Bider, 1998).
Turtles that nest on seasonally saturated
substrates, such as sandbars, are of special
interest regarding the potential impact of
atrazine because many are already imperiled
(Moll and Moll, 2004) and possess attributes
of their natural history that might render them
especially susceptible to xenobiotic substances. In late spring and early summer, when
atrazine is usually applied to agricultural
fields, these turtles construct subterranean
terrestrial nests in which the eggs are subject
to a wide range of environmental conditions
that impact embryonic development (Deeming, 2004). Most of these species produce
flexible-shelled eggs, which readily exchange
moisture with the surrounding soil in the nest
during incubation (Packard et al., 1987).
These eggs also have the capacity to incorporate larger molecules across the eggshell that
can influence the phenotypes of developing
embryos (Wibbels et al., 1991a), rendering
them good models for chemicals such as
endocrine disruptors that show effects at low
ATRAZINE (2-chloro-4-ethythlamino-6-isopropylamine-1,3,5-tiazine) is one of the most
widely applied herbicides in the United States
and the most commonly used herbicide in the
world (Murphy, 2005). Between 64 and 75
million pounds of atrazine are used annually
in the United States, most commonly in
agricultural fields in the Midwest (Nations
and Hallberg, 1992; USEPA, 2009). Applied
as a nonselective preemergent chemical for
control of broadleaf plants, degradation-resistant atrazine is subject to meteorological
factors that distribute it more broadly than
intended, causing it to appear in most sources
of water, including rain (Brent et al., 2001).
Emerging research indicates that lower
concentrations of atrazine can cause more
damage both behaviorally and physiologically
to the organism than higher concentrations
(Crews et al., 2000; Hayes et al., 2002a;
Propper, 2005; Storrs and Kiesecker, 2004).
However, the majority of atrazine research
has focused on amphibians and fish, and little
is known about the effects of atrazine on other
wild vertebrates, such as reptiles (Rohr and
2
PRESENT ADDRESS: Department of Biology, Utah State
University, Logan, UT 84322, USA
3
CORRESPONDENCE: email, lorin215@gmail.com
23
24
HERPETOLOGICA
doses (Willingham and Crews, 2000). Moreover, most turtles have temperature-dependent sex determination (TSD; Ewert et al.,
1994; Janzen and Paukstis, 1991), whereby the
gonads and sex-related structures are shaped
by a hormonal milieu controlled by temperatures experienced during the middle portion
of embryonic development (Crews et al.,
1994; Wibbels et al., 1991b). Furthermore,
the posthatching behavior of neonatal turtles
is critical for their survival during a time of
intense predation (Janzen et al., 2000b), yet
xenobiotic substances such as atrazine may
affect the ability to escape predation, feeding
behaviors, and other motivational behaviors
(Guillette and Gunderson, 2001; Clotfelter et
al., 2004). Thus, sex determination and other
chemically mediated traits in turtles might be
especially susceptible to abiotic influences
such as atrazine that impact the developing
endocrine system.
To test for phenotypic effects of atrazine on
embryonic and neonatal turtles, we obtained
eggs from sandbar nests of two common,
widely distributed North American river turtles: the Ouachita map turtle (Graptemys
ouachitensis) and the false map turtle (G.
pseudogeographica). These species are not of
conservation concern, but they can serve as
models for species that are imperiled, including
their endangered congeners, because they have
flexible-shelled eggs and TSD. We incubated
eggs under controlled conditions in the laboratory, by using standard levels for sand
moisture and developmental temperature,
and values for atrazine concentrations that
reflect levels recorded in the study area
(Kalkhoff et al., 2000). As outcomes, we
assessed variation in hatching success, incubation time, shell abnormalities, body size, the
ability to leave a surrogate nest-like structure,
sex, time required to right when overturned,
and time to swim to the surface—for hatchlings
in the treatment and control groups. We also
reared a subset of neonates for 11 mo after
hatching to evaluate variation among treatment
and control groups in growth, survival, escape
behavior, time to first foraging event, and time
to catching a prey item. Our findings raise
significant concerns regarding the phenotypic
and potential fitness effects of atrazine on
nontarget organisms such as riverine turtles.
[Vol. 67, No. 1
MATERIALS AND METHODS
Egg Collection and Incubation
We collected 106 eggs in early June from 10
nests that were laid within the previous 24 h
(six nests of G. ouachitensis and four nests of
G. pseudogeographica) on a major sandbar in
the Cedar River in the vicinity of Wiese
Slough near Muscatine, Iowa, USA. This
sandbar is regularly inundated in late spring
and early summer (FJJ, personal observation)
by water carrying agricultural runoff from
upstream, including significant concentrations
of atrazine (Iowa Department of Natural
Resources, 2000). After retrieval from the
nests, we immediately transported the eggs to
Iowa State University in Styrofoam chests
containing sand from the nests.
We placed eggs from each nest into the
control group (1 mL of acetone to 1 L of
dechlorinated, deionized water) or one of the
three atrazine treatment groups (0.1, 1.0, and
100.0 mg/L of atrazine in the same acetone:
water vehicle added to the incubation substrate) that bracketed the concentrations of
atrazine detected in the river near the field
site (Iowa Department of Natural Resources,
2000). We used acetone to aid solvency of the
atrazine (assessed at .99% purity by gas
chromatography by Ultra Scientific). We
randomly chose two eggs per nest for each
control and treatment group (i.e., eight eggs
total per nest). We placed one egg from each
nest in a plastic box for incubation, such that
each box contained 10 eggs. Thus, two boxes
represented each control and treatment
group. The remaining 26 eggs were placed
immediately at 220uC.
We created the hydric conditions in each
box by adding 31.02 g of the relevant control
or treatment liquid to 943 g of heat-sterilized
‘‘playground’’ sand, yielding a moist substrate
for incubation. We chose this starting condition based on weighing a sample of sand from
the excavated nests, drying the sand, and
calculating the percentage that evaporated
(mean 5 3.28% liquid). We placed all boxes in
an environmental chamber set at a constant
28uC, which is the incubation temperature
that naturally produces a somewhat malebiased sex ratio in Graptemys (Vogt and Bull,
1984), to enhance the probability of detecting
any feminizing effects of the atrazine treat-
March 2011]
HERPETOLOGICA
ments. We rotated boxes within the environmental chamber twice weekly to minimize the
possible effect of thermal gradients on the
developing embryos. After the original application of the control or treatment mixture, we
measured water loss and rehydrated each box
once weekly with dechlorinated, deionized
water to maintain a relatively constant concentration of substrate moisture. This approach was intended to simulate a one-time
inundation or exposure of freshly laid eggs to
atrazine-containing water, as might be experienced under natural conditions for riverine
turtles, and not to maintain constant atrazine
concentration.
Offspring Assessment
We recorded the time between laying and
hatching as a measure of incubation time. If a
turtle hatched from an egg and survived at
least one day, we considered this a successful
hatching event. We assigned to each turtle
that successfully hatched a unique number
with a permanent marker on the carapace for
identification. We also clipped the marginal
scutes in a unique pattern as a more
permanent form of marking. Within 24 hours
after hatching, we visually examined each
turtle for carapace abnormalities, defined as a
deviation from 12 left and 12 right marginal
scutes, 4 left and 4 right pleural scutes, and/or
5 vertebral scutes (Ernst and Lovich, 2009).
We also weighed the turtles (6 0.01 g) on an
electronic balance and obtained carapace
length and width (6 0.01 mm) with digital
calipers. We recorded these size measurements again for all turtles that survived for
11 mo after hatching.
Initial Performance Tests
Performance tests, which occurred in the
first week after hatching, assessed the ability
of hatchlings in righting and swimming trials.
We performed all tests at 22uC. In a righting
trial, we flipped a turtle onto its carapace to
simulate an avian predation event and recorded the time taken to right itself (and thereby
avoid predation; Ashmore and Janzen, 2003).
If a turtle took .10 min, we terminated the
test and recorded a time of 600 s. After we
timed a turtle in righting, we placed it directly
into a container of water, with a break of no
25
more than 30 s between the righting trial and
the swimming trial.
For swimming trials, we gently pushed a
turtle to the bottom of a plastic shoebox
containing 8.5 cm of dechlorinated, deionized
water. We measured the time (in seconds)
from release of a turtle until the moment
when the turtle’s snout broke the surface of
the water. We tested each turtle only once,
and all turtles were tested in the same water.
When the initial righting and swimming
tests were completed, we killed approximately
one turtle from each clutch in all control and
treatment groups by pericardial injection of
0.5 mL of a 1:1 mixture of sodium pentobarbital:deionized water. We opened these 36
sacrificed turtles and one dead full-term
embryo to examine gross gonadal morphology
(Gutzke and Paukstis, 1984; Yntema, 1981).
We noted gonadal sex and observations of
unusual features of related structures for each
turtle, particularly cortical tissue on testes and
residual oviducts in males. All sacrificed
turtles then received a museum tag and were
stored in 70% ethanol as voucher specimens.
Rearing and Long-Term Behavior Tests
We reared the 36 remaining turtles in
individual containers with dechlorinated, deionized water at 23uC with full-spectrum light
on a 12 h:12 h on:off cycle for 4 mo. Initially,
each turtle received one piece of ReptominE
(average length, 15 mm) without the researcher present. After 10 min, the researcher
returned and noted which turtles had eaten.
We removed and discarded untouched food,
but left partially consumed food in the
container. We placed a new piece of Reptomin in the container if the food was completely consumed. Once every turtle had
eaten at least one time, we placed individuals
from each treatment group in a communal
container (approximately 4 mo after hatching).
We fed these turtles to excess two or three
times per week.
To assess ability to escape a nest, we housed
hatchlings individually in containers (17.3 cm
long 3 12.6 cm wide 3 6.1 cm high, with
2.5 cm of dechlorinated, deionized water) for
4 mo. Every 3 d, we noted the individuals that
had escaped from their containers to a larger
common area (1.5 3 0.5 m) and returned
26
HERPETOLOGICA
[Vol. 67, No. 1
TABLE 1.—Incubation time, hatching success, scute abnormalities, and sex ratio of hatchling Graptemys
pseudogeographica and G. ouachitensis turtles as a function of atrazine treatment. The control and treatment groups
started with 20 eggs each.
Variable
Control
0.1 mg/L Atrazine
10.0 mg/L Atrazine
100.0 mg/L Atrazine
Hatching success (%)
Incubation time
(days 6 SE)
Scute abnormalities (%)
95
58.6 6 0.64
(n 5 19)
25
(n 5 19)
89
(n 5 19)
85
57.8 6 0.50
(n 5 17)
17
(n 5 19)
94
(n 5 19)
90
58.4 6 0.46
(n 5 18)
11
(n 5 18)
78
(n 5 18)
90
59.0 6 0.44
(n 5 18)
0
(n 5 18)
72
(n 5 18)
Sex ratio (% male)
these escaped turtles to their original containers. We quantified escape behavior by comparing the number of successful escapes for
the 30 total trials over the 4 mo.
Foraging Ability
At approximately 10 mo after hatching, we
assessed each neonate’s ability to capture live
invertebrate prey. We placed a turtle in a 10gal. glass aquarium filled with 7.5 cm of water.
We then put a small cricket (Acheta) in the
water, initiating the trial. We measured the
time the turtle took to capture the cricket, at
which point we stopped the trial. When the
single cricket had been present for 25 min
(1500 s), we put another cricket in the water to
ensure that the turtle was exposed to an active
cricket. If a turtle did not successfully strike at
either cricket after 45 min (2700 s), we then
removed the turtle. When all turtles had been
tested, we returned the turtles to their original
containers and fed them Reptomin.
Sex and Gonadal Observation of
Remaining Neonates
Other than turtles that died during the
experimental period, we similarly sacrificed,
sexed, and preserved as voucher specimens
the remaining turtles shortly after 11 mo of
posthatching growth, behavior, and survival
assessments.
Statistical Analyses
We assessed possible effects of atrazine on
the dependent variables statistically by comparing all treatment and control groups. We
used nominal logistic tests, which require a 0,
1 count (e.g., live, dead) for the dependent
variables, to analyze hatching success (live,
dead), scute abnormalities (presence, absence), sex (male, female), and survival to
11 mo (live, dead) among the control and
treatment groups; survival analyses would not
have provided more insight because nearly all
the mortality was concentrated in one treatment. We evaluated data for incubation time,
the three measures of body size (weight,
length, and width), the five measures of
performance (righting time, swimming time,
number of escapes, time to first feeding, and
foraging ability), and the three measures of
posthatching growth to 11 mo (mass, length,
and width) with analyses of covariance among
the control and treatment groups. Five
measures were not normally distributed and
therefore were transformed before statistical
analysis: we log transformed righting time,
swimming time, time to first feeding, and
foraging ability (food capture time), and
square-root transformed number of escapes.
Egg mass at oviposition was the potential
covariate for incubation time and measures of
body size at hatching, whereas mass at hatching
was the covariate for the measures of performance and posthatching growth. Species, the
species-by-atrazine-treatment interaction, and
clutch nested within species were independent
variables in all statistical tests, with the latter
variable considered a random effect. Statistical
analyses were performed using JMP, version
6.0 (SAS Institute, 2005).
RESULTS
Embryonic Development
Of the 80 eggs that began incubation in this
experiment, 72 (90%) hatched. Eggs hatched
between 6 and 13 August, approximately 8–
9 wk after oviposition; nine offspring exhibited
scute abnormalities, with six of them having
extra marginal scutes and the remainder
having abnormal vertebral scutes (Table 1).
March 2011]
HERPETOLOGICA
27
TABLE 2.—Morphometrics of hatchling Graptemys pseudogeographica and G. ouachitensis turtles as a function of
atrazine treatment. All values are given means 6 SE.
Variable
Initial
Egg mass (g)
Mass (g)
Carapace length (mm)
Carapace width (mm)
At 11 mo
Mass (g)
Carapace length (mm)
Carapace width (mm)
Control
11.24
(n
8.44
(n
31.91
(n
28.64
(n
12.70
(n
40.29
(n
37.80
(n
6
5
6
5
6
5
6
5
6
5
6
5
6
5
0.34
20)
0.26
19)
0.33
19)
0.30
19)
0.80
9)
0.91
9)
0.96
9)
The sex ratio (both the hatchlings initially
sexed and those sexed later) was 61 males and
11 females (Table 1). Eight males exhibited
traces of ovarian cortex on their testes
(individuals sexed immediately) and seven
males showed signs of a residual oviduct
(individuals sexed later). Except for sex (P 5
0.72), clutch made a substantial contribution
to the other four variables (P , 0.04 in all
cases), including explaining nearly 70% of the
variance in incubation time. Scute abnormalities declined as atrazine values increased
(Table 1). For all other embryonic development traits, we identified no statistically
significant differences among the control and
treatment groups, between the two species, or
with respect to species-by-treatment interactions (P . 0.05 in all cases).
Offspring Size and First-Year Growth
At hatching, turtles from different treatment groups did not differ significantly in
mass, carapace length, or carapace width
(Table 2). Although initial egg mass explained
significant variation in body size at hatching (P
, 0.0003 in all cases), none of the three body
size variables at hatching contributed significantly to body size at 11 mo (P . 0.80 in all
cases). Only carapace length at hatching
differed between the two species (P 5 0.02),
with G. ouachitensis averaging approximately
1 mm longer than G. pseudogeographica.
With the exception of mass at hatching
(73.4%), clutch contributed minimally to body
size (,28% in all cases). No measures of size
0.1 mg/L atrazine
10.0 mg/L atrazine
11.12 6 0.30
(n 5 20)
8.38 6 0.27
(n 5 18)
31.85 6 0.42
(n 5 18)
28.75 6 0.37
(n 5 18
13.70 6 0.86
(n 5 5)
41.57 6 0.93
(n 5 5)
38.94 6 0.86
(n 5 5)
11.13
(n
8.40
(n
31.78
(n
28.52
(n
12.10
(n
39.55
(n
37.67
(n
6
5
6
5
6
5
6
5
6
5
6
5
6
5
0.34
20)
0.27
18)
0.41
18)
0.32
18)
1.21
8)
1.59
8)
1.47
8)
100.0 mg/L atrazine
11.12
(n
8.39
(n
31.82
(n
28.84
(n
12.68
(n
40.57
(n
38.62
(n
6
5
6
5
6
5
6
5
6
5
6
5
6
5
0.33
20)
0.26
18)
0.37
18)
0.35
18)
0.80
10)
1.33
10)
0.95
10)
at hatching or after 11 mo of growth differed
significantly among control and atrazine-treated offspring (Table 2), or as a consequence of
a species-by-treatment interaction (P . 0.15
in all cases).
Performance and First-Year Survival
Righting and swimming times were not
significantly correlated (r 5 20.15, P 5 0.32,
n 5 45). Righting times, swimming times, and
foraging abilities were statistically indistinguishable among the control and treatment
groups (Table 3) and with respect to speciesby-treatment interactions (P . 0.15 in all
cases). Smaller turtles at hatching began
foraging sooner than larger offspring. Turtles,
especially heavier turtles, from control eggs
escaped more frequently from their containers than did hatchlings from the lowest and
highest treatment groups (Table 3). Overall,
these results indicate that turtles from the
control group generally were more physically
active than turtles from the atrazine-treated
groups both shortly after hatching and many
months later. These behavioral differences
seem to be reflected in the significant overall
reduction in first-year survival of hatchlings
from the 0.1 mg/L treatment compared with
hatchlings over that same period from the
control and other treatment groups (Table 3).
DISCUSSION
The global environment is increasingly
laden with anthropogenically derived chemicals and their breakdown products. Many of
28
HERPETOLOGICA
[Vol. 67, No. 1
TABLE 3.—Offspring performance as a function of atrazine treatment for Graptemys pseudogeographica and G.
ouachitensis. Except for survival to 11 mo, values are least squares means 6 SE. Species was a significant source of
variation only for swimming time (P 5 0.03). Values for clutch describe the percentage of total variance explained by this
random effect.
Variable
Righting time (s)
(n 5 45)
Swimming time (s)
(n 5 72)
Time to 1st feeding (days)
(n 5 36)
Escapes (n 5 36)
Foraging time (s) (n 5 32)
Survival (%)
0.1 mg/L
Atrazine
10.0 mg/L
Atrazine
100.0 mg/L
Atrazine
180 6 50.1
109 6 48.9
178 6 56.6
169 6 47.2
202 6 65.4
260 6 69.0
345 6 67.1
396 6 67.1
69 6 11.5
85 6 12.9
72 6 12.7
86 6 11.6
2.9 6 0.39
1.6 6 0.44
3.0 6 0.44
1.9 6 0.39
37 6 100.9
135 6 140.4
300 6 108.3
39 6 96.4
90
63
100
100
Control
these compounds, including atrazine-containing products, have been implicated as adversely affecting a variety of organisms. In this
experiment, we explored the possible impacts
of environmentally relevant levels of atrazine
under ecologically realistic conditions on a
suite of fitness-related traits in neonatal
freshwater turtles. Most notably, we found
that turtles exposed as embryos to the lowest
level of atrazine (0.1 mg/L) in the incubation
substrate had significantly decreased posthatching survival in the laboratory.
Reptiles can display abnormalities in their
reproductive system (Crain et al., 1999;
Guillette et al., 1994), sex steroid profiles
(Willingham et al., 2000), liver physiology
(Ganser et al., 2003), swimming performance
(Hopkins and Winne, 2006), and secondary
sex characteristics (e.g., sexual dimorphism;
de Solla et al., 2002) in response to contaminants. The most substantial impact of contaminants such as endocrine disruptors is
usually noted in individuals that are exposed
early during development. Studies focusing
specifically on atrazine and related herbicides
found that embryos exhibited effects long
after exposure that relate directly to survival,
behavior, and reproduction (Bigsby et al.,
1999; Clotfelter et al., 2004; Rohr and McCoy,
2010). Indeed, we observed no deleterious
effects of atrazine treatment on any trait
measured at hatching, but we found a
significant effect on abnormalities in carapace
scute numbers, for which we have no
Treatment
effect
F
P
F
P
F
P
F
P
F
P
x2
P
5
5
5
5
5
5
5
5
5
5
5
5
0.27
0.84
1.39
0.25
0.83
0.49
2.09
0.13
0.65
0.59
16.27
0.001
Clutch
effect (%)
0
11.2
27.8
3.7
25.0
0
explanation, and a deleterious effect on
several posthatching traits (see below).
Despite the presumed sensitivity of gonadal
development in animals with TSD to endocrine disruptors, such as atrazine, we found no
detectable impact of our atrazine treatments
on offspring sex ratio or on feminization of
neonatal male map turtles (Graptemys) in this
study. We did not address this question at the
ultrastructural level, because lack of an effect
at the gross morphological level was not
encouraging. Our finding contrasts with results for slider turtles (Trachemys scripta;
Willingham, 2005), but it is consistent with
observations on common snapping turtles
(Chelydra serpentina; de Solla et al., 2006),
both species with TSD. This variation in
outcomes is perhaps not surprising given the
differences in methodology for atrazine application between our experiment and Willingham’s study and the concordance in
methodology for exposing eggs to atrazine in
our experiment and de Solla et al.’s study.
In the wild, Graptemys eggs take 2–3 mo to
hatch, with young generally remaining approximately 10 cm below the surface in the
sandy nests for several days until the yolk sac
is absorbed (Vogt, 1980). After this period,
neonates usually emerge from the nests and
embark on a many-meter trek to reach a
freshwater environment where they are subject to numerous (Vogt, 1980). Thus, individual fitness is substantially impacted at several
key points during these early life stages.
March 2011]
HERPETOLOGICA
Clearly, though, the first challenge to a
hatchling’s survival ability is to emerge successfully from the nest. If the hatchling is
unable to escape, it will die in the nest (Peters
et al., 1994) or be subject to parasitism (Vogt,
1981). Although not quite statistically significant, the results of our study show that
hatchlings treated with the lowest concentration of atrazine (0.1 mg/L) were least capable
of escaping from their nest-like enclosures.
This diminished capacity was accompanied by
reduced abilities of neonates from atrazinetreated eggs compared with control turtles for
a suite of performance traits likely related to
fitness in the wild (Table 3). The relative
consistency of the findings suggests that the
results could be confirmed with larger sample
sizes.
Our most dramatic results relate to firstyear survival. Mortality in the wild during this
period is substantial, deriving from a series of
abiotic (e.g., low temperatures; Costanzo et
al., 2008) and biotic (e.g., predation; Janzen et
al., 2000a) factors. We raised neonates in
benign posthatching conditions where such
factors were eliminated, yet still observed
significantly reduced first-year survival of
turtles deriving from eggs treated with 0.1 mg/
L atrazine. Although competition for food is
unlikely, factors possibly involved in the
mortality include disease and physiological
abnormalities (Forson and Storfer, 2006; Rohr
et al., 2008). In nature, atrazine-induced
mortality is particularly problematic for longlived species, such as turtles. If enough young
fail to reach sexual maturity (4–8 yr in our
study species; Vogt, 1980) to replace reproductive individuals that die, populations can
experience bottlenecks, inbreeding depression, or both (Kuo and Janzen, 2004; Újvári
et al., 2002). Moreover, this delayed maturity
could hinder research efforts to assess the
long-term effects of endocrine disruptors,
such as atrazine.
None of our results exhibited a rising
impact with increasing concentration of atrazine. Instead, the control group and highest
treatment group showed similar effects for
several of the parameters, a pattern that has
been noted previously with atrazine and other
endocrine disruptors. These chemicals often
display the greatest biological effects at lower
29
concentrations and in a nonmonotonic dose–
response curve (Hayes et al., 2002b, 2003;
Hayes, 2005; Storrs and Keisecker, 2004;
Willingham, 2005). Our results were consistent with these and other studies in that the
treatment group that deviated the most from
the control group in escape behavior and firstyear survival received the lowest concentration of atrazine (i.e., 0.1 mg/L) shortly after
oviposition.
Our study has limitations that may influence the conclusions. Beyond relatively small
sample sizes, we did not measure atrazine in
sand from which eggs were collected, in eggs
at oviposition, or in hatchlings. Consequently,
we do not know the amounts of atrazine to
which eggs were exposed before the experiment, nor do we know about the rate or
mechanism of incorporation of atrazine into
the embryos and hatchlings. However, because eggs from each clutch were randomly
assigned to treatments and all clutches were
represented in all treatments, any prior
exposure or other clutch effects should not
influence the among-treatment results that we
obtained.
We also applied atrazine just once, at the
beginning of development. If applied at a
different time or throughout the incubation
period, the effects on traits such as sex ratio
may have been different (sensu Willingham,
2005). Many studies have focused on the
results of chronic exposure to atrazine, but
such exposure may be more relevant to
amphibians than to terrestrial vertebrates.
The fact that we obtained meaningful results
for some important traits with only a one-time
application early in development therefore
raises concern regarding the biological impacts of atrazine specifically and other xenobiotic substances in general. Our findings
provide evidence that organisms could be
adversely affected after only minimal exposure
to low concentrations of atrazine.
Acknowledgments.—We thank R. Paitz for assistance
collecting the eggs, A. Bronikowski for help with
formulating the atrazine stock solutions, the Janzen lab,
the Mullin lab (Eastern Illinois University), and S. de
Solla and anonymous reviewers for helpful comments on
the manuscript. Eggs were collected under scientific
collector permit SC 14 to FJJ from the Iowa Department
of Natural Resources, and turtles were handled in
accordance with approved protocol 5-03-5457-J from the
30
HERPETOLOGICA
Committee on Animal Care at Iowa State University. This
work was funded by National Science Foundation grants
IBN-0080194 and DEB-0089680, and analysis and writing
were completed while the authors were supported by
DEB-0640932 and REU supplement DEB-0822673.
LITERATURE CITED
ASHMORE, G. M., AND F. J. JANZEN. 2003. Phenotypic
variation in smooth softshell turtles (Apalone mutica)
from eggs incubated in constant versus fluctuating
temperatures. Oecologia 134:182–188.
BIGSBY, R., R. E. CHAPIN, G. P. DASTON, B. J. DAVIS, J.
GORSKI, L. E. GRAY, K. L. HOWDESHELL, T. R. ZOELLER,
AND F. S. vOM SAAL. 1999. Evaluating the effects of
endocrine disruptors on endocrine function during
development. Environmental Health Perspective Supplements 107:613–618.
BRENT, R. N., J. SCHOFIELD, AND K. MILLER. 2001. Results
of the Lake Michigan mass balance study: Atrazine data
report. Available at: http://www.epa.gov/greatlakes/
lmmb/results/atra_datarpt.html. US Environmental
Protection Agency.
CLOTFELTER, E., A. M. BELL, AND K. R. LEVERING. 2004.
The role of animal behaviour in the study of endocrine
disrupting chemicals. Animal Behaviour 68:465–476.
COSTANZO, J. P., R. E. LEE, JR., AND G. R. ULTSCH. 2008.
Physiological ecology of overwintering in hatchling
turtles. Journal of Experimental Zoology 309A:297–379.
CRAIN, D. A., I. D. SPITERI, AND L. J. GUILLETTE, JR. 1999.
The functional and structural observations of the
neonatal reproductive system of alligators exposed in
ovo to atrazine, 2,4-D, or estradiol. Toxicology and
Industrial Health 15:180–185.
CREWS, D., J. M. BERGERON, J. J. BULL, D. FLORES, A.
TOUSIGNANT, J. K. SKIPPER, AND T. WIBBELS. 1994.
Temperature-dependent sex determination in reptiles:
Proximate mechanism, ultimate outcomes, and practical
applications. Developmental Genetics 15:297–312.
CREWS, D., E. WILLINGHAM, AND J. K. SKIPPER. 2000.
Endocrine disruptors: Present issues, future directions.
Quarterly Review of Biology 75:243–260.
DE SOLLA, S. R., C. A. BISHOP, AND R. J. BROOKS. 2002.
Sexually dimorphic morphology of hatching snapping
turtles (Chelydra serpentina) from contaminated and
reference sites in the Great Lakes and St. Lawrence
River Basic, North America. Environmental Toxicology
and Chemistry 21:922–929.
DE SOLLA, S. R., P. A. MARTIN, K. J. FERNIE, B. J. PARK,
AND G. MAYNE. 2006. Effects of environmentally
relevant concentrations of atrazine on gonadal development of snapping turtles (Chelydra serpentina).
Environmental Toxicology and Chemistry 25:520–526.
DEEMING, D. C. 2004. Reptilian Incubation: Environment, Evolution and Behaviour. Nottingham University
Press, Nottingham, UK.
ERNST, C. H., AND J. E. LOVICH. 2009. Turtles of the
United States and Canada, 2nd Edition. Johns Hopkins
University Press, Baltimore, Maryland, USA.
EWERT, M. A., D. R. JACKSON, AND C. E. NELSON. 1994.
Patterns of temperature-dependent sex determination
in turtles. Journal of Experimental Biology 270:3–15.
FORSON, D., AND A. STORFER. 2006. Atrazine increases
Ranavirus susceptibility in the tiger salamander,
[Vol. 67, No. 1
Ambystoma
tigrinum.
Ecological
Applications
16:2325–2332.
GANSER, L. R., W. A. HOPKINS, L. NEIL, S. HASSE, J. H.
ROE, AND D. M. SEVER. 2003. Liver histopathology of
the southern Watersnake, Nerodia fasciata fasciata,
following chronic exposure to trace element-contaminated prey from a coal ash disposal site. Journal of
Herpetology 37:219–226.
GUILLETTE, L. J., JR., AND M. P. GUNDERSON. 2001.
Alteration in development of reproductive and endocrine systems of wildlife populations exposed to
endocrine-disrupting contaminants. Reproduction
122:857–864.
GUILLETTE, L. J., JR., T. S. GROSS, G. R. MASSON, J. M.
MATTER, H. F. PERCIVAL, AND A. R. WOODWARD. 1994.
Developmental abnormalities of the gonad and abnormal sex hormone concentrations in juvenile alligators
from contaminated and control lakes in Florida.
Environmental Health Perspectives 102:680–688.
GUTZKE, W. H. N., AND G. L. PAUKSTIS. 1984. A low
threshold temperature for sexual differentiation in the
painted turtle. Copeia 1984:546–547.
HAYES, T. B. 2005. Welcome to the revolution: Integrative
biology and assessing the impact of endocrine disruptors on environmental and public health. Integrative
and Comparative Biology 45:321–329.
HAYES, T. B., A. COLLINS, M. LEE, M. MENDOZA, N.
NORIEGA, A. A. STUART, AND A. VONK. 2002a. Hermaphroditic, demasculinized frogs after exposure to the
herbicide atrazine at low ecologically relevant doses.
Proceedings of the National Academy of Science USA
99:5476–5480.
HAYES, T. B., K. HASTON, M. TSUI, A. HOANG, C.
HAEFFELE, AND A. VONK. 2002b. Herbicides: Feminization of male frogs in the wild. Nature 419:895–896.
HAYES, T. B., K. HASTON, M. TSUI, A. HOANG, C.
HAEFFELE, AND A. VONK. 2003. Atrazine-induced
hermaphroditism at 0.1 ppb in American leopard frogs:
Evidence from the laboratory and the wild. Environmental Health Perspectives 111:1–8.
HOPKINS, W. A. 2000. Reptile toxicology: Challenges and
opportunities on the last frontier in vertebrate ecotoxicology. Environmental Toxicology and Chemistry
19:2391–2393.
HOPKINS, W. A., AND C. T. WINNE. 2006. Influence of body
size on swimming performance of four species of
neonatal natricine snakes acutely exposed to a cholinesterase-inhibiting pesticide. Environmental Toxicology
and Chemistry 25:1208–1213.
IOWA DEPARTMENT OF NATURAL RESOURCES. 2000. Section
305(b) Water Quality Report. Available at: http://wqm.
igsb.uiowa.edu/WQA/305b/2000/2000_305b.html.
JANZEN, F. J., AND G. L. PAUKSTIS. 1991. Environmental sex
determination in reptiles: Ecology, evolution, and
experimental design. Quarterly Review of Biology
66:149–179.
JANZEN, F. J., J. K. TUCKER, AND G. L. PAUKSTIS. 2000a.
Experimental analysis of an early life-history stage:
Avian predation selects for larger body size of hatchling
turtles. Journal of Evolutionary Biology 13:947–954.
JANZEN, F. J., J. K. TUCKER, AND G. L. PAUKSTIS. 2000b.
Experimental analysis of an early life-history stage:
Selection on size of hatchling turtles. Ecology
81:2290–2304.
March 2011]
HERPETOLOGICA
KALKHOFF, S. J., K. K. BARNES, K. D. BECHER, M. E.
SAVOCA, D. J. SCHNOEBELEN, E. M. SADORF, S. D.
PORTER, AND D. J. SULLIVAN. 2000. Water quality in the
eastern Iowa basins, Iowa and Minnesota, 1996–98. US
Geological Survey Circular 1210. Available at: http://
pubs.water.usgs.gov/circ1210/.
KUO, C.-H., AND F. J. JANZEN. 2004. Genetic effects of a
persistent bottleneck on a natural population of ornate
box turtles (Terrapene ornata). Conservation Genetics
5:425–437.
MOLL, D., AND E. MOLL. 2004. The Ecology, Exploitation,
and Conservation of River Turtles. Oxford University
Press, New York, New York, USA.
MURPHY, M. 2005. Atrazine banned in EU but safe in U.S.
Chemistry and Industry 3:10.
NATIONS, B., AND G. HALLBERG. 1992. Pesticides in Iowa
precipitation. Journal of Environmental Quality
21:486–492.
PACKARD, G. C., M. J. PACKARD, K. MILLER, AND T. J.
BOARDMAN. 1987. Influence of moisture, temperature
and substrate on snapping turtle eggs and embryos.
Ecology 68:983–993.
PETERS, A., K. J. F. VERHOEVEN, AND H. STRIJBOSCH. 1994.
Hatching and emergence in the Turkish Mediterranean
loggerhead turtle, Caretta caretta: Natural causes for
egg and hatchling failure. Herpetologica 50:369–373.
PROPPER, C. 2005. The study of endocrine-disrupting
compounds: Past approaches and new directions.
Integrative and Comparative Biology 45:194–200.
ROHR, J. R., AND K. A. MCCOY. 2010. A qualitative metaanalysis reveals consistent effects of atrazine on
freshwater fish and amphibians. Environmental Health
Perspectives 118:20–32.
ROHR, J. R., A. M. SCHOTTHOEFER, T. R. RAFFEL, H. J.
CARRICK, N. HALSTEAD, J. T. HOVERMAN, C. M. JOHNSON,
L. B. JOHNSON, C. LIESKE, M. D. PIWONI, P. K. SCHOFF,
AND V. R. BEASLEY. 2008. Agrochemicals increase
trematode infections in a declining amphibian species.
Nature 455:1235–1239.
SAS INSTITUTE. 2005, JMP, version 6.0. SAS Institute,
Cary, North Carolina, USA.
SAUMURE, R. A., AND J. R. BIDER. 1998. Impact of
agricultural development on a population of wood
turtles (Clemmys insculpta) in southern Quebec,
Canada. Chelonian Conservation and Biology 3:37–45.
SOLOMON, K. R., J. A. CARR, L. H. DU PREEZ, J. P. GIESY,
R. J. KENDALL, E. E. SMITH, AND G. J. VAN DER KRAAK.
2008. Effects of atrazine on fish, amphibians, and
31
aquatic reptiles: A critical review. Critical Reviews in
Toxicology 38:721–772.
STORRS, S. I., AND J. M. KIESECKER. 2004. Survivorship
patterns of larval amphibians exposed to low concentrations of atrazine. Environmental Health Perspectives
112:1054–1057.
ÚJVÁRI, B., T. MADSEN, T. KOTENKO, M. OLSSON, R. SHINE,
AND H. WITTZELL. 2002. Low genetic diversity threatens
imminent extinction for the Hungarian meadow viper
(Vipera ursinii rakosiensis). Biological Conservation
105:127–130.
USEPA (US ENVIRONMENTAL PROTECTION AGENCY). 2009.
Atrazine science re-evaluation: Potential health risks.
Available at: http://www.regulations.gov/search/Regs/
home.html#documentDetail?R50900006480a3dabc.
VOGT, R. C. 1980. Natural history of the map turtles
Graptemys pseudogeographica and G. ouachitensis in
Wisconsin. Tulane Studies in Zoology and Botany
22:17–48.
VOGT, R. C. 1981. Turtle egg (Graptemys: Emydidae)
infestation by fly larvae. Copeia 1981:457–459.
VOGT, R. C., AND J. J. BULL. 1984. Ecology of hatchling sex
ratio in map turtles. Ecology 65:582–587.
WIBBELS, T., J. J. BULL, AND D. CREWS. 1991a. Synergism
between temperature and estradiol: A common pathway in turtle sex determination. Journal of Experimental Zoology 260:130–134.
WIBBELS, T., J. J. BULL, AND D. CREWS. 1991b. Temperature-dependent sex determination: A mechanistic
approach. Journal of Experimental Zoology 270:71–78.
WILLINGHAM, E. 2005. The effects of atrazine and
temperature on turtle hatchling size and sex ratios.
Frontiers in Ecology and the Environment 3:309–313.
WILLINGHAM, E., AND D. CREWS. 2000. The red-eared
slider turtle: An animal model for the study of low doses
and mixtures. American Zoologist 40:421–428.
WILLINGHAM, E., T. RHEN, J. SAKATA, AND D. CREWS. 2000.
Embryonic treatment with xenobiotics disrupts steroid
hormone profiles in hatchling red-eared slider turtles
(Trachemys scripta elegans). Environmental Health
Perspectives 108:329–333.
YNTEMA, C. L. 1981. Characteristics of gonads and
oviducts in hatchlings and young of Chelydra serpentina resulting from three incubation temperatures.
Journal of Morphology 167:297–304.
.Accepted: 26 September 2010
.Associate Editor: William Lutterschmidt