Access to this full-text is provided by Wiley.
Content available from Functional Ecology
This content is subject to copyright. Terms and conditions apply.
Functional Ecology. 2023;37:2895–2909.
|
2895wileyonlinelibrary.com/journal/fec
Received: 19 May 2023
|
Accepted: 31 July 2023
DOI : 10.1111/136 5-243 5.14 420
RESEARCH ARTICLE
Differential early- life survival underlies the adaptive
significance of temperature- dependent sex determination in a
long- lived reptile
Samantha L. Bock1,2 | Yeraldi Loera3 | Josiah M. Johnson1,2 |
Christopher R. Smaga1,2 | David L. Haskins2,4 | Tracey D. Tuberville2 | Randeep Singh5 |
Thomas R. Rainwater5,6 | Philip M. Wilkinson6 | Benjamin B. Parrott1,2
1Eugene P. Odum Sch ool of Ecology, University of Georg ia, Athe ns, Georgia, USA; 2Sava nnah River Ecolog y Laborator y, Aiken, South Car olina, USA;
3Depar tment of Ecology & Evolutionary B iolog y, Princeton Univer sity, Princeton, N ew Jers ey, USA; 4Warne ll School of Fores try & Natural Resources,
University of Georgia , Athens , Georgia, USA ; 5Belle W. Baruch Ins titute of Coast al Ecology & Forest Science, Clemson Universit y, Georgetown, South Carolina,
USA and 6Tom Yawkey Wild life Center, Georgetown, South Carolina, USA
This is an op en access arti cle under the ter ms of the Creative Commons Attribution-NonCommercial License , which permits use, dis tribu tion and reprod uction
in any medium, provided the original work is properl y cited an d is not use d for comm ercial purposes.
© 2023 The Authors . Functional Ecology published by John Wiley & Sons Ltd on behal f of British Ecological So ciety.
Correspondence
Samantha L. Bo ck
Email: samantha.bock@uga.edu
Benjamin B. Parrott
Email: benparrott@srel.uga.edu
Funding information
Odum Sch ool of Ecol ogy; National Science
Foundation, Grant/Award Number:
1754903; Department of Ener gy Of fice of
Environmental Management, Grant/Award
Number: DE- EM0005228
Handling Editor: Pau Carazo
Abstract
1. Many ectotherms rely on temperature cues experienced during development to
determine offspring sex. The first descriptions of temperature- dependent sex
determination (TSD) were made over 50 years ago, yet an understanding of its
adaptive significance remains elusive, especially in long- lived taxa.
2. One novel hypothesis predicts that TSD should be evolutionarily favoured when
two criteria are met— (a) incubation temperature influences annual juvenile
survival and (b) sexes mature at different ages. Under these conditions, a sex-
dependent effect of incubation temperature on offspring fitness arises through
differences in age at sexual maturity, with the sex that matures later benefiting
disproportionately from temperatures that promote juvenile survival.
3. The American alligator (Alligator mississippiensis) serves as an insightful model in
which to test this hypothesis, as males begin reproducing nearly a decade after
females. Here, through a combination of artificial incubation experiments and
mark- recapture approaches, we test the specific predictions of the survival- to-
maturity hypothesis for the adaptive value of TSD by disentangling the effects of
incubation temperature and sex on annual survival of alligator hatchlings across
two geographically distinct sites.
4. Hatchlings incubated at male- promoting temperatures (MPTs) consistently
exhibited higher survival compared to those incubated at female- promoting
temperatures. This pattern appears independent of hatchling sex, as female s pro-
duced from hormone manipulation at MPT exhibit similar survival to their male
counterparts.
2896
|
Functional Ecology
FUNCTIONAL ECOLOGY
1 | INTRODUC TION
The existence of distinct sexes is a fundamental feature of nearly
all metazoan taxa, however the systems by which these sexes
arise are remarkably diverse (Bachtrog et al., 2014; Capel, 2017).
Many taxa lack sex chromosomes and instead rely on environ-
mental cues experienced during discrete periods in development
to determine offspring sex (Bull, 1980 ; Devlin & Nagahama, 2002;
Hobaek & Larsson, 199 0). While a range of environmental factors
operate in systems of environmental sex determination (ESD), in-
cluding pH, nutrition, photoperiod, and social context (reviewed
in Korpelainen, 19 90 ), the most taxonomic ally widespread form of
ESD is temperature- dependent sex determination (TSD; Bull, 1980 ;
Capel, 2017 ). Despite the first descriptions of TSD being made over
50 years ago (Charnier, 1966), an understanding of its adaptive sig-
nificance has remained elusive.
Work by Charnov and Bull (1977 ) pioneered early thinking re-
garding the evolutionary underpinnings of TSD. Consistent with ex-
isting theory rooted in frequency- dependent selection, their model
also integrated aspects of conditional sex allocation theory, with the
central idea that TSD should be favoured when reproductive females
have limited control over the environment their offspring will enter
and there is a sex- dependent effect of temperature on offspring
fitness (i.e. temperature- by- sex interaction; Charnov & Bull, 1977;
Shine, 1999). When these conditions are met, TSD represent s an
adaptive sex allocation strategy which allows females to preferen-
tially produce the sex that will benefit most from the incubation
conditions they experience. Though considerable research effort
has focused on exploring sex- dependent effects of the develop-
mental thermal environment on subsequent traits related to repro-
ductive fitness, empirical support for such effects remains scarce.
In the Atlantic silverside (Menidia menidia), females are produced at
low temperatures characteristic of the early breeding season while
males are produced at warmer temperatures characteristic of the
late breeding season (Conover, 1984; Conover & Kynard, 1981).
This temperature- linked difference in hatch timing leads to marked
sexual size dimorphism due to females' extended growing season.
Accordingly, adult fecundity in females is more highly dependent
on body size than it is in males resulting in a sex- specific effect of
developmental temperature on adult fecundity (Conover, 1984). The
first strong empirical evidence for the adaptive value of TSD in an
amniote ver tebrate came with a study in the jacky dragon, Amphi-
bolurus muricatus (Warner & Shine, 2008). Using hormonal manipu-
lations to produce males and females across a range of temperatures
and quantifying lifetime reproductive success in a semi- natural field
enclosure, the authors demonstrated that reproductive success of
both males and females is maximized at the temperatures that nor-
mally produce each respective sex (Warner & Shine, 2008).
Demonstrating a sex- dependent effect of the developmental
environment on adult fecundity in long- lived species has proven
more difficult. Beyond challenges associated with measuring com-
ponents of fitness for long- lived species in the field, seasonal effects
on of fspring growth like those observed in the silverside and jacky
dragon are unlikely to translate into differences in reproduc tive per-
formance when individuals do not reach maturity for several years
or even decades. This may indicate that TSD is adaptively neutral
in these cases (Sabath et al., 2016; Valenzuela & Lance, 2004) or
that adult fecundity is not the relevant target of selection (Sæther
et al., 2013). Effects of the developmental environment on the other
key component of fitness, survival, have received comparatively
less attention. While many studies have documented temperature
effects on organismal traits presumably linked to sur vival, including
growth, morphology, and behaviour (reviewed in Noble et al., 2018),
few have quantified survival in the wild. This is especially pertinent
in the context of a novel hypothesis put forth by Schwanz and col-
leagues which suggests TSD is evolutionarily favoured when two
criteria are met— (1) incubation temperature influences juvenile
survival regardless of sex, and (2) sexes mature at different ages
(Schwanz et al., 2016). Under these conditions, a sex- dependent
effect of temperature on fitness arises through differences in age
at sexual maturity, with the sex that matures later benefiting dis-
proportionately from incubation temperatures that confer higher
probability of juvenile survival. Consistent with this ‘survival- to-
maturity’ hypothesis, species with TSD tend to exhibit greater sex
differences in age at maturity than those in species with genotypic
sex determination (Bókony et al., 2019; Katona et al., 2021; Schwanz
et al., 2016). However, a strong empirical test of this hypothesis in a
field context is lacking.
5. Additional experiments show that incubation temperature may affect early- life
survival primarily by affecting the efficiency with which maternally transferred
energy resources are used during development.
6. Results from this study provide the first explicit empirical support for the adap-
tive value of TSD in a crocodilian and point to developmental energetics as a
potential unifying mechanism underlying persistent survival consequences of in-
cubation temperature.
KEY WORDS
evolution, reptile, survival, temperature- dependent sex determination
|
2897
Functional Ecology
BOCK et al.
The American alligator ( Alligator mississippiensis) provides a par-
ticularly compelling system in which to assess evidence for survival-
to- maturity as a potential mechanism underlying the adaptive value
of TSD. Alligator embryos exhibit robust phenotypic responses to
subtle changes in incubation temperature (Bock et al., 2021), and
extensive nest temperature monitoring suggests maternal nesting
behaviour is limited in its capacity to influence incubation tempera-
tures (Bock et al., 2020). Further, while both female and male alliga-
tors are physiologically capable of reproducing upon reaching ~1.8 m
in total leng th, paternity analyses demonstrate most males only
begin siring offspring in the wild after reaching sizes much larger
than this (e.g. >2.8 m; Zajdel et al., 2019), lead in g to a st ark sex di ffe r-
ence in age at first reproduction (~14 years in females vs. ~24 years
in males; Wilkinson et al., 2016). Given these observations, males
would be predicted to benefit more from incubation temperatures
conferring greater juvenile survival compared to females. Thus,
male- promoting temperatures (MPTs) should confer greater early-
life sur vival compared to female- promoting temperatures (FPTs) if
differential survival- to- maturity underlies the adaptive significance
of TSD in this species.
The present study aimed to test the specific predictions of the
survival- to- maturity hypothesis for the adaptive value of TSD in the
American alligator. Toward this end, we pursued three central re-
search questions: (1) Does incubation temperature influence early-
life survival of alligator hatchlings, and specifically do MPTs confer
increased survival? (2) If so, is this due to a direct influence of in-
cubation temperature or an effect of hatchling sex? And (3) what
mechanisms mediate the lasting influence of incubation temperature
on post- hatching survival? A preliminar y test of potential influences
of incubation temperature and/or sex on early- life survival in alli-
gators was conducted by characterizing alligator population sex ra-
tios across different size classes based on previously published field
capture data (Table S1). This meta- analytical approach was followed
by experiments employing a combination of artificial incubation
treatments and hormone manipulations to disentangle the effects
of temperature from those of sex on a suite of metabolic and or-
ganismal phenotypes (Figure 1). Early- life sur vival of hatchlings was
subsequently quantified through mark- recapture approaches. Taken
together, results from this study provide the first explicit empirical
support for the adaptive value of TSD in a crocodilian.
2 | MATERIALS AND METHODS
2.1 | Summary of population sex ratios across size
classes of the American alligator
Effect s of sex and/or incubation temperature on early- life sur-
vival rates are predicted to yield differences in sex ratios across
size classes. Variation in American alligator population sex ratios
across size classes was characterized using published data from 16
field studies of juvenile and adult size classes summarized in (Lance
et al., 2000), and four studies of hatchlings and/or nest temperatures
(Table S1). Size classes were defined based on thresholds for key life
history transitions (Ferguson, 1985). Studies including a size class
that spanned multiple size class definitions in the present study were
excluded from analysis. Sex ratios of each size class were summa-
rized across the species' geographic range by taking the mean sex
ratio of studies from each state (North Carolina = NC, South Caro-
lina = SC, Louisiana = LA, Florida = FL) weighted by the reported sam-
ple size. Sexing hatchlings remains challenging in this species due to
the lack of sexually dimorphic morphology in early life stages (Bock
et al., 2022), thus empirically derived hatchling sex ratios were sup-
plemented with predicted sex ratios from nest temperatures. Nest
temperatures were translated to predicted hatchling sex ratios using
the established temperature- by- sex ratio reaction norm for the
American alligator (Lang & Andrews, 1994) and mean nest tempera-
tures measured during the thermosensitive period, the window of
time during which sex responds to temperature (Bock et al., 2020).
2.2 | Field collections and incubation experiments
All experimental procedures were approved by the Institutional
Animal Care and Use Committee of the University of Georgia and
field collections were permitted by the South Carolina Depar tment
of Natural Resources (SC- 08- 2019, SC- 08- 2020, and SC- 08- 2021).
Eggs were collected from wild alligator nest s across three consecu-
tive reproduc tive seasons. In June 2019, nests were located via air-
boat at Par Pond, a 1120 ha freshwater reser voir on the Department
of Energy's Savannah River Site (PAR; Aiken, SC). Four clutches of
eggs (n = 148) were collected prior to the canonical thermosensitive
period of sex determination (Ferguson stage 20– 24). In June of 2020
and 2021, nest s were located via helicopter aerial surveys and sub-
sequently accessed on foot at the Yawkey Wildlife Center (Y WC;
Georgetown, SC). In both 2020 and 2021, eight clutches of eggs
(2020: n = 372; 2021: n = 413) were collected prior to Ferguson stage
14. All eggs were transported back to the Savannah River Ecology
Laboratory (Aiken, SC) in their natal nesting material. Within 12 h of
arrival at the laboratory, a representative embryo from each clutch
was staged according to (Ferguson, 1985). All eggs were candled to
assess viability, weighed, and transferred to damp sphagnum moss
where they were maintained at 32°C in programmable incubator
chambers (Percival Scientific, model I36NLC).
Experimental incubation treatments implemented in 2019,
2020, and 2021 built upon one another in a stepwise manner (Fig-
ure 1; Table 1). To test whether hatchlings incubated at a FPT and
MPT dif fered in their early- life sur vival, in 2019 eggs were distrib-
uted between two constant incubation temperatures, either 29°C
(FPT; predicted 100% female) or 33.5°C (MPT; predicted ~89%
male), at Ferguson stage 17 and were maintained at these tem-
peratures until hatching. To test whether observed differences in
early- life sur vival between hatchlings incubated at FPT and MPT
were due to incubat ion temp erature or hatchling sex , in 2020 eggs
were first distributed between t wo constant temperatures, either
29.5°C (FPT) or 33.5°C (MPT ) at stage 15, and then received an
2898
|
Functional Ecology
FUNCTIONAL ECOLOGY
exogenous dose of either 17β- estradiol (E2; 0.5 μg/g egg weight;
Sigma Aldrich, E2758) or vehicle control (VEH; 0.5 μL/g eg g weight
absolute ethanol) at stage 19. This dose of 17β- estradiol was cho-
sen because it has been shown to induce complete sex reversal
at MPT (Kohno et al., 2015). Thus, this 2 × 2 factorial design in-
cluded treatment groups resulting in both presumptive males
(MPT- VEH) and presumptive females from the MPT (MPT- E2),
thereby disentangling incubation temperature from sex. Finally,
to test how incubation at a high- female promoting temperature
(HFPT) influenced early- life survival relative to that of hatchlings
from the lower incubation temperatures, in 2021 eggs were dis-
tribute d at stage 15 between t hre e con stant temperatures, 29.5°C
(FPT), 33.5°C (MPT) or 34.5°C (HFPT; predicted 34% female), and
were maintained at these temperatures until hatching. It should
be noted that temperatures above 34°C tend to produce variable
sex ratios which can be male- or female- biased, however, for the
purposes of this study we refer to 34.5°C as a ‘high- female pro-
moting temperature’ to highlight its distinction from temperatures
FIGURE 1 (a) Graphical summar y of survival- to- maturity hypothesis for the adaptive value of temperature- dependent sex determination.
(b) Temperature- by- sex ratio reaction norm of the American alligator and corresponding predictions for influences of incubation temperature
on early- life survival. (c) Schematic of experimental designs implemented across study years with predicted sex ratios for each treatment
group and corresponding sample sizes (number of hatchlings released for mark- recapture). FPT (+E2), female promoting temperature
(29.5°C) with addition of 17β- estradiol (0.5 μg/g egg weight); FPT (control), female promoting temperature (29°C in 2019, 29.5°C in 2020,
2021) with either no topical treatment or addition of vehicle control (0.5 μL /g egg weight absolute ethanol; 2020 only); HFPT (control), high
female- promoting temperature (34.5°C) with no topical treatment; MPT (+E2), male- promoting temperature (33.5°C) with addition of 17β-
estradiol (0.5 μg/g egg weight); MPT (control), male- promoting temperature (33.5°C) with either no topical treatment or addition of vehicle
control (0.5 μL/g egg weight absolute ethanol; 2020 only).
(a) (b)
(c)
Scale of inference
Scale at which the
factor of interest is
applied Number of replicates at the appropr iate scale
Individuals Individuals 2019: 38, 62
2020: 41, 40, 54, 58
2021: 33, 51, 30
TAB LE 1 Replication statement.
|
2899
Functional Ecology
BOCK et al.
that produce consistently male- biased sex ratios. Apart from the
34.5°C treatment, the other incubation treatments were chosen
because these temperatures reliably produce highly male- biased
or fe ma le- bia sed hatch ling sex r atios. Whil e the experime nt al tem-
perature treatments do not reflect the thermal complexity of nat-
ural nests (e.g. daily fluctuations), mean temperatures of alligator
nests previously measured at YWC have been shown to encompass
all treatment temperatures implemented here (Bock et al., 2020).
Onset (UTBI- 001) HOBO temperature loggers preprogrammed to
record temperature at 5 min intervals were kept in the substrate
adjacent to eggs to ensure experienced temperatures matched the
intended experimental temperatures. Across all experiments, the
timing at which each clutch reached key developmental stages was
predicted based on the established relationship between incuba-
tion temperature and developmental rate in this species (Kohno &
Guillette, 2013).
2.3 | Embryonic respirometry trials
To determine how incubation temperature influences developmen-
tal energetics in the American alligator, embryonic respirometry
trials were conducted for a subset of individuals (n = 56) in 2020.
Res pi ro metr y t rials were conducted at st age 26 , wh ic h occu rs after
sex determination and the approximate developmental timepoint
at which metabolic rate peaks in Crocodylus johnstoni (Whitehead
& Seymour, 199 0). Trials were conducted using a flow- through
respirometry system (Field Metabolic System, FMS; Sable Instru-
ments) at the same temperature at which eggs were incubated, ei-
the r 29.5°C or 33 .5°C. Eg gs were weigh ed just pr ior to the trial an d
then were placed in individual plastic metabolic chambers (473 mL),
each with an inflow and outflow channel. Metabolic chambers
and an empty control chamber were kept within the FMS meta-
bolic co oler and the constant tempera ture through out the trial was
controlled by a programmable PELT- 5 device (Sable Instruments).
A constant flowrate of 50 mL/min was used for all trials. All tri-
als began between 0945 and 2200 h. Eggs were placed in their re-
spective metabolic chambers and allowed to acclimate within the
FMS for 1 h prior to the initiation of measurement. Trials included
one to three eggs and each egg's metabolic chamber was meas-
ured sequentially for 25 min, with transitions between chambers
controlled by an automated multiplexer (Sable Instruments). Con-
secutive metabolic measurements were separated by a 5 min meas-
urement of the baseline control chamber and all trials ended with a
10 min baseline measurement to allow for later correction of sensor
drift over the course of the trial. Data from the respirometr y tri-
als were recorded using ExpeDat a software (version 1.7.30; Sable
Systems). Previously published custom scripts (Stager et al., 2021;
https://github.com/Mstag er/batch_proce ssing_Exped ata_files)
implemented in R statistical software version 4.1.2 were used to
correct for drift in baseline O2 levels and extract minimum oxy-
gen consumption (VO2) averaged over a 10 min period from the raw
metabolic data. Embryonic VO2 measurements were subsequently
used to approximate the energetic cost of development by taking
the produc t of embryonic metabolic rate (EMR) and incubation
duration as proposed by (Marshall et al., 2020). While this metric
of developmental cost does not account for changes in metabolic
rate over the course of development, it does capture relative dif-
ferences between incubation temperatures.
2.4 | Mark- release- recapture methods
Each individual was weighed and snout- vent leng th (SVL), total
length, and tail girth (circumference of tail at vent) were measured
upon hatching. Hatchlings were individually identified by clipping
a unique pattern of keratinous tail scutes and housed together in
a temperature- controlled indoor facility in custom fibreglass tanks
that allowed individuals to swim and bask freely (Bock et al., 2021).
Hatchlings were not fed during this period. When hatchlings reached
9– 14 days old, individuals were haphazardly assigned to pods of
8– 26 hatchlings each and transpor ted back to their site of origin.
Pods of hatchlings were released at a single location within ~350 m
of one another at their site of origin (PAR or YWC) over the course
of 3 weeks. Release locations were chosen based on the availability
of suitable habitat for hatchlings (e.g. presence of permanent fresh-
water and aquatic vegetation to provide cover) and accessibility for
subsequent recapture efforts.
Monthly rec apture effort s commenced at least 2 weeks after the
last release date and consisted of exhaustive searches occurring on
one or two consecutive night s between the hours of 1730 and 0200.
Hatchlings were located visually via eyeshine (Subalusky, 2007) and
were caught by hand or net from a canoe or by researchers on foot.
The search area was defined as within ~50 m of the shoreline and no
further than ~250 m from a release site. Previous studies suggest alli-
gator hatchlings generally do not disperse more than ~200 m during
their first year of life (Deitz, 1979). Recapture efforts proceeded
until hatchlings could no longer be located or captured within the
search area. Recapture effor ts occurred an average of 4 weeks apart,
excluding the winter dormancy period (December– early March)
during which hatchlings were assumed to be inactive (Deitz, 1979).
The length of time bet ween the last pre- winter recapture effort and
first post- winter recapture ef fort for 2019, 2020 and 2021 was 26,
19 and 20 weeks, respectively. Monthly recapture efforts continued
through the first year post- hatch.
2.5 | Data processing and statistical analyses
Statistical analyses were conducted in R version 4.1.2 (R Core
Tea m , 2021), unless indicated otherwise. To test for differences in
alligator population sex ratios across size classes, published sex ratio
(Table S1) was coded as an individual binary response variable based
on reported proportion male and sample size (i.e. number of males
[1] and females [0]) and modelled using a binomial generalized linear
mixed effects model (GLMM) with a logit link function in the ‘lme4’
2900
|
Functional Ecology
FUNCTIONAL ECOLOGY
package (Bates et al., 2015). Hatchling sex ratios predicted from
nest temperatures were excluded from this analysis. The candidate
model explaining variation in sex included a fixed effect of size class
and a random effect of study identity. Coefficients of this GLMM are
reported as log- odds ratios (±SE), with the adult size class ser ving as
the reference level.
Analysis of data from the artificial incubation experiments was
separated into three phases. The first phase of analysis aimed at
testing whether incubation temperature influences hatchling sur-
vival. Survival was assessed via two approaches. First, survival sta-
tus was defined based on whether an individual was recaptured in
the present time period or any subsequent time period. Survival
in the pre- winter and post- winter periods were separately treated
as binomial response variables and modelled using GLMMs with a
logit link function in the ‘lme4’ package (Bates et al., 2015). Candi-
date models explaining variation in sur vival included fixed effects
of incubation temperature, hormone treatment (2020 only) and pre-
sumptive sex (2020 only) as well as random effects of clutch and
pod identity. Coefficients of the sur vival GLMMs are repor ted as
log- odds ra tios (±SE), wi th FPT ser vi ng as the ref erence le vel in te m-
perature comparisons for all cohorts.
Survival was also assessed by modelling individual capture his-
tories with s ta nd ard Cormack- Jolly- Se be r (C JS) mo dels in Program
MAR K (version 10.0) to generate max imum like lihood est imate s of
both apparent sur vival (Φ) and recapture probability (p; White &
Burnham, 1999). The use of the CJS model served to test whether
observed influences of incubation temperature on survival status
were due to biased recapture rates rather than apparent survival
differences. Capture histories encompassed recapture efforts
from October through July of the following year. Each year's co-
hort was modelled separately and a weekly timestep was imple-
mented to account for unequal sampling intervals. For all cohort s,
incubation temperature was treated as a categorical grouping
variable and season (pre- winter/winter, post- winter) was treated
as a time- dependent covariate. For the 2020 cohort, presumptive
sex was treated as an individual covariate. Models were fit using
a logit- link function and candidate models were compared based
on Akaike's information criterion adjusted for small sample sizes
(AICc). For the 2019 and 2021 cohor ts, four candidate models
were compared which variably included temperature effects on
Φ and/or p. For the 2020 cohort, 16 candidate models were com-
pared which variably included effects of temperature and/or pre-
sumptive sex on Φ and/or p. An effect of season on both Φ and/or
p was included in all candidate models.
The second phase of analysis aimed at testing whether incu-
bation temperature and/or sex reversal via exogenous oestrogen
treatment influences hatchling morphometric traits. The depen-
dent variables SVL , hatchling mass and BMI (mass/[2 × SVL]) were
each modelled in the ‘lme4’ package using LMMs with fixed ef-
fects of incubation temperature, hormone treatment (2020 only),
presumptive sex (2020 only) and egg mass as well as a random
effect of clutch identity. The third phase of analysis aimed at test-
ing whether temperature- dependent hatchling morphometric
traits explain variation in survival status. Similar to the approach
described previously, survival in the pre- winter and post- winter
periods were t reated as binomia l response variables and modelled
using GLMMs with a logit link function, however in this case, can-
didate models included different combinations of fixed effects of
hatchling morphometric traits previously shown to respond to in-
cub ation tem pe rature. Pri or to fittin g thes e mode ls , ea ch trait was
rescaled to have a mean of zero and standard deviation of one.
Model coefficients of these LMMs are reported for the rescaled
pr edi c to r s. All ca ndi dat e m ode ls in c lu d ed ran dom ef f ect s of cl utc h
and pod identit y.
Finally, to address the extent to which incubation temperature
influences the energetic cost of development, the dependent vari-
ables incubation duration (days between predicted oviposition date
and pip date), EMR (VO2), and developmental cost (product of in-
cubation duration and EMR; Marshall et al., 2020) were modelled
with LMMs including fixed effects of incubation temperature and
egg mass, and a random effect of clutch identit y.
Across all analyses (apart from those implemented in program
MARK), models including subsets of fixed effects were compared
to the global model based on AICc using the package ‘AICcmodavg’
(Mazerolle, 2020). Models with ∆AICc <2.0 were considered to have
support. If the best model for a response variable included a predic-
tor with more than two levels, post- hoc comparisons between lev-
els were conducted with the package ‘emmeans’ (Lenth et al., 2023)
and p- values were adjusted according to Tukey's HSD method. For
LMMs, degrees of freedom for post- hoc comparisons were calcu-
lated according to the Kenward- Roger method. Any deviations
of the global model from model assumptions (e.g. overdispersion)
were diagnosed via the ‘simulateResiduals’ function in the package
‘DHARMa’ (Hartig, 2022).
3 | RESULTS
3.1 | Population sex ratios become increasingly
male- biased in older size classes
Population sex ratios across the geographic range of the Ameri-
can alligator show a consistent trend with nearly balanced or
female- biased sex ratios in hatchlings and marked male biases
observed in juvenile and adult size classes (Figure 2). Indeed, the
best model explaining variation in alligator sex included a fixed
effect of size class (βHatchling = −1.80 ± 0.24; βJuvenile = −0.42 ± 0.05;
Table S2) and post- hoc comparisons indicated significant differ-
ences in the probability of an individual being male between all
size classes (p < 0.0001). In the two studies which determined the
sex of over 6000 naturally incubated hatchlings using genital mor-
pholog y, yearl y sex ratios range d from 10.6% to 42.4% male (Elsey
& Lang, 2014; Rhodes & Lang, 1996). Further, mean nest tem-
peratures during the thermosensitive period in Florida and South
Carolina also were predicted to yield female- biased hatchling sex
ratios (40.6% and 30.6% male, respectively). In contrast, juvenile
|
2901
Functional Ecology
BOCK et al.
sex ratios were male- biased across each of the states for which
data were available. This was also true for adult sex ratios, with the
exception of Florida (Figure 2).
3.2 | Hatchlings from a MPT exhibit higher survival
than those from a FPT
Over the course of the study period, 407 hatchlings were released
and 115 unique individuals were recaptured at least once. Hatch-
lings incubated at MPT showed higher survival in both the pre-
winter and post- winter periods compared to hatchlings incubated at
FPT (Figure 3). In 2019, survival, as measured by recapture status,
was ~1.8 times higher for MPT hatchlings in the pre- winter period
and ~6.7 times higher in the post- winter period (Figure 3a). For
each cohort, the GLMM that best explained variation in pre- winter
survival included a fixed effect of incubation temperature (2019:
βMPT = 0.91 ± 0.46; 2020: βMPT = 2.18 ± 0.78; 2021: βMPT = 1.19 ± 0.57;
Figure 3a; Table S3). However for the 2021 cohort, the null model
of pre- winter sur vival was within 1.0 ∆AICc of the top model
(∆AICc = 0.45, wi = 0.44). The best model explaining variation in post-
winter survival for the 2019 and 2020 cohor ts also included only
a fixed effect of incubation temperature (2019: βMPT = 2.08 ± 1.07;
2020: βMPT = 3.48 ± 1.34; Figure 3a; Table S4).
In the case of the CJS models, the top model for both 2019 and
2021 included an effect of incubation temperature on apparent sur-
vival but not recapture probability (2019: βMPT = 1.23 ± 0.49; 2021:
βMPT = 1.28 ± 0.32; Figure 3b; Table 2). In contrast, the top CJS model
for the 2020 cohort included an effect of incubation temperature on
recapture probability, but not apparent survival (βMPT = 3.77 ± 0.56;
Table 2). Still, the model including an effect of incubation tempera-
ture on both apparent survival and recapture probability was within
1.0 ∆AICc (Φ: βMPT = 0.53 ± 0.45, p: βMPT = 3.07 ± 0.82; Figure 3b;
Table 2).
3.3 | Incubation at MPT promotes hatchling
survival independent of sex
Control male and sex- reversed female hatchlings from the MPT
showed similar survival, as measured by recapture status, in both the
pre- winter (Proportion surviving: MPTControl = 0.35, MPTE2 = 0.36)
and post- winter periods (Proportion surviving: MPTControl = 0.22,
MPTE2 = 0.26; Figure 3a). The GLMM that best explained variation
FIGURE 2 Patterns of American alligator sex ratio variation across size classes. (a) State- level weighted mean sex ratio for juvenile and
adult size classes (adults defined by total length greater than 1.8 m). Darker shaded regions correspond to the weighted mean maximum and
minimum total length for each size class. Lighter shaded regions correspond to the absolute maximum and minimum total length included in
each size class. Actual and predicted hatchling sex ratios are depicted as a single point for each individual study. (b) Sex ratios by size class
across the Americ an alligator geographic range. Each pie chart depicts a sex ratio reported by an individual study. Pie charts are associated
with the locations at which each study took place. FL, Florida; LA, Louisiana; NC , North Carolina; SC, South Carolina, USA.
(a)
(b)
r
2902
|
Functional Ecology
FUNCTIONAL ECOLOGY
in pre- winter and post- winter survival retained only a fixed effect
of incubation temperature while excluding the fixed effects of pre-
sumptive sex and hormone treatment (Figure 3a; Tables S3 and S4).
In addition, neither of the top- performing CJS models for the 2020
cohort included an effect of sex on apparent survival or recapture
probability (Table 2). Collectively, these results suggest observed
survival dif ferences between hatchlings from the MPT and hatch-
lings from the FPT are due to direct effects of incubation tempera-
ture rather than a confounding influence of sex. Further, embryonic
oestrogen exposure does not appear to incur any consequences for
hatchling survival.
3.4 | Incubation at a high FPT confers reduced
hatchling survival
Based on the shape of the alligator temperature- by- sex ratio reac-
tion norm and the predictions of the survival- to- maturity hypothe-
sis, high incubation temperatures that can promote the development
of females (>34°C; HFPT) should confer reduced sur vival compared
to those incubation temperatures that produce highly male- biased
sex ratios (33.5°C; MPT; Figure 1b). Indeed, in the 2021 cohort, a
lower proportion of hatchlings from the HFPT survived in the pre-
winter (0.27) and post- winter (0.03) periods compared to hatchlings
FIGURE 3 Incubation treatment effec ts on pre- and post- winter hatchling survival. (a) Proportion of hatchlings with known sur vival
status in the pre- winter and post- winter periods. Known survival status was defined by an individual being recaptured at least once in
the present time period or a subsequent time period. (b) Weekly apparent survival estimates from Cormack- Jolly- Seber models including
effects of both temperature and season on apparent survival (Φ) and recapture probability (p). Vertical lines depict ±1 SE. FPTcontrol, female
promoting temperature (29°C in 2019, 29.5°C in 2020, 2021) with either no topical treatment or addition of vehicle control (0.5 μL/g egg
weight absolute ethanol; 2020 only); FPTE2, female promoting temperature (29.5°C) with addition of 17β- estradiol (0.5 μg/g egg weight);
HFPTcontrol, high female- promoting temperature (34.5°C) with no topical treatment; MPTcontrol, male- promoting temperature (33.5°C)
with either no topical treatment or addition of vehicle control (0.5 μL/g egg weight absolute ethanol; 2020 only); MPTE2, male- promoting
temperature (33.5°C) with addition of 17β- estradiol (0. 5 μg/g egg weight).
(a)
(b)
|
2903
Functional Ecology
BOCK et al.
from the MPT (pre- winter: 0.39, post- winter: 0.16; Figure 3a). How-
ever, while the GLMM that best explained variation in pre- winter
survival of the 2021 cohort included a fixed effect of incubation
temperature, post- hoc comparisons suggest this is likely driven by
the difference in sur vival between hatchlings from the MPT and
FPT, as survival differences between hatchlings from the HFPT and
MPT (log- odds ratio = −0.37, p = 0.78) and HFPT and FPT (log- odds
ratio = 0.82, p = 0.45) were comparatively weaker. No hatchlings
from the FPT were recaptured in the post- winter period for the 2021
cohort thereby limiting parameter estimation.
3.5 | Hatchlings incubated at MPT tend to be larger
than those incubated at FPT
Incubation treatments not only contributed to variation in
hatchling survival, but also to variation in hatchling morphol-
ogy. Hatchling SVL was best explained by the LMM includ-
ing fixed effects of both incubation temperature and egg mass
for the 2019 (βMPT = 0.11 ± 0.07, β
eggmass = 0.04 ± 0.01) and 2021
(βMPT = 0.23 ± 0.08, βHFPT = −0.69 ± 0.10, β
eggmass = 0.02 ± 0.003;
Table S5) cohorts. Hatchlings from the MPT tended to have a longer
SVL than hatchlings from other incubation temperatures (Fig-
ure S1A). Post- hoc comparisons of the three temperatures in the
2021 cohort confirmed this trend— MPT hatchlings were longer than
both HFPT hatchlings (p < 0.0 001) and FPT hatchlings (p = 0.03).
FPT hatchlings were also longer than HFPT hatchlings (p < 0.0001).
Incubation temperature was also associated with SVL in the 2020
cohort (βMPT = −0.30 ± 0.04), however, this effect was in the oppo-
site direction of other years with FPT hatchlings being longer than
MPT hatchlings (Figure S1A). Interestingly, oestrogen treatment in
addition to egg mass was also included in the top model for SVL
(βE2 = 0.19 ± 0.04, β
eggmass = 0.03 ± 0.0 03; Table S5), with oestro-
gen treatment associated with longer hatchlings. SVL was the only
hatchling trait for which the ef fect of incubation temperature was
inconsistent across cohorts.
For all cohorts, the best model explaining variation in both hatch-
ling mass and BMI included effects of incubation temperature and
egg mass, with hatchlings from the MPT exhibiting consistently larger
body mass (2019: βMPT = 3.78 ± 0.52, β
eggmass = 0.47 ± 0.07; 2020:
βMPT = 1.38 ± 0.35, β
eggmass = 0.45 ± 0.04; 2021: βMPT = 4.86 ± 0.65,
βHFPT = 0.59 ± 0.77, β
eggmass = 0.49 ± 0.06) and higher BMI (2019:
βMPT = 0.14 ± 0.02, β
eggmass = 0.02 ± 0.003; 2020: βMPT = 0.11 ± 0.01,
β
eggmass = 0.02 ± 0.002; 2021: βMPT = 0.17 ± 0.02, βHFPT = 0.14 ± 0.03,
β
eggmass = 0.02 ± 0.002) compared to hatchlings from the FPT (Fig-
ure S1B,C; Tables S6 and S7). Post- hoc comparisons of the three
incubation temperatures in 2021 showed that while hatchlings
from the MPT were larger in mass than hatchlings from both the
FPT (p < 0.00 01) and HFPT (p < 0.0001), hatchlings from the FPT
and HFPT did not differ from each other (p = 0.73). In the case of
BMI, however, hatchlings from the MPT and HFPT showed higher
BMI than hatchlings from the FPT (p < 0.0001), but BMI did not dif-
fer between the MPT and HFPT (p = 0.42). Presumptive sex was not
included in the top model for any of the traits examined. Overall,
temperature consistently exerted a strong influence on hatchling
morphology, and specifically, incubation at MPT promoted the de-
velopment of larger hatchlings with greater body condition (as indi-
cated by BMI). Given this observation, further analyses were aimed
at determining whether these temperature- related traits might ex-
plain variation in hatchling sur vival.
TAB LE 2 Comparison of Cormack- Jolly- Seber models for 2019, 2020 and 2021 cohorts. All candidate models shown for 2019 and 2021
cohorts. Top three of 16 candidate models based on Akaike's information criterion adjusted for small sample sizes (AICc) and null model
shown for 2020 cohort.
Cohort Rank
Model
KAICc∆AICcwiLikelihoodΦp
2019 1Tinc + season Season 5300.68 0.00 0.68 1.00
2Tinc + season Tinc + season 6302.84 2.16 0.23 0.34
3Season Tinc + season 5305.68 4.99 0.06 0.08
Null Season Season 4307.07 6.39 0.03 0.04
2020 1Season Tinc + season 5475.90 0.00 0.32 1.00
2Tinc + season Tinc + season 6476.75 0.85 0.21 0.66
3Season Tinc + sex + season 647 7.9 1 2.01 0.12 0. 37
… … … … … … … …
Null Season Season 4533.18 57. 28 0 .11 0.00
2021 1Tinc + season Season 6298.35 0.0 0 0. 74 1.00
2Tinc + season Tinc + season 8301.23 2.88 0.17 0 .24
3Season Tinc + season 6302.58 4.23 0.09 0.1 2
Null Season Season 4311.19 12.83 0.00 0.00
Abbreviations: K, model parameter s; p, recapture probability; season, pre- winter or post- winter period; Tinc, incubation temperature; wi, model
weight; Φ, apparent sur vival.
2904
|
Functional Ecology
FUNCTIONAL ECOLOGY
3.6 | Survival is weakly associated
with temperature- dependent hatchling
morphometric traits
In the 2019 and 2021 cohorts, pre- winter survival was best explained
by variation in hatchling SVL , wherein longer hatchlings were more
likely to sur vive (2019: βSVL = 0.44 ± 0.23; 2021: βSVL = 0.56 ± 0.30;
Figure 4a; Table S8). The second- best model of pre- winter sur vival
in the 2019 cohort included a single effect of body mass, with larger
hatchlings again showing a survival advantage over smaller hatch-
lings (βMASS = 0.37 ± 0.21; Figure 4b; Table S8). Nonetheless, for
both cohorts, the null model was within 2 ∆ AICc of the top models
(Table S8). Further, post- winter survival was not well explained by
any of the temperature- related traits examined. Results were simi-
larly mixed for the 2020 cohort. While pre- winter survival was not
well explained by any hatchling traits, the top model for post- winter
FIGURE 4 Associations between pre- winter survival status and temperature- related hatchling traits. Pre- winter survival status with
respect to (a) hatchling snout- vent length (SVL), (b) body mass and (c) body mass index (BMI). Solid line depict s the fit of a generalized linear
model of pre- winter survival status with a binomial error distribution and each morphological trait. Dotted lines represent ±1 SE. ‘ALL’
column includes hatchlings from all three cohor ts pooled. FPT, female promoting temperature; HFPT, high female- promoting temperature;
MPT, male- promoting temperature.
(a)
(b)
(c)
|
2905
Functional Ecology
BOCK et al.
survival included an ef fect of BMI, with higher condition hatchlings
exhibiting greater sur vival (Figure 4c; Table S9). While a few asso-
ciations between survival and temperature- related hatchling mor-
phometric traits were detected, no single trait emerged as a strong
candidate for mediating the effect of incubation temperature on
survival.
3.7 | Energetic cost of development is minimized at
male promoting temperatures
Incubation temperature may alternatively exert lasting effects on
hatchling survival via influences on physiological characteristics not
ne ces s a rily re f lecte d in mo r p h olog y suc h as po s t- ha t chin g en e rg y re-
serves. Variation in metabolic rate was best explained by temperature
and egg mass (βMPT = −0.0004 ± 0.0002, β
eggmass = 0.0008 ± 0.0 001),
though the temperature effect was not in the expected direction and
the effect size was relatively small (Figure 5a; Table S10). Incubation
duration, on the other hand, was strongly associated with tempera-
ture (βMPT = −13.14 ± 0.33), with embryos at MPT hatching ~13 days
earlier than those at FPT (Figure 5b; Table S10). As a result, deve lop -
mental cost, quantified as the product of EMR and incubation dura-
tion (Marshall et al., 2020), was strongly influenced by temperature
(βMPT = −3.01 ± 0.21, β
eggmass = 0.09 ± 0.02; Figure 5c; Table S10).
Embryos at FPT incurred a greater energetic cost of development,
largely due to increases in incubation duration (Figure 5d).
4 | DISCUSSION
Incubation temperature exert s a strong influence on the subsequent
survival of hatchling alligators, with temperatures promoting male
development conferring greater early- life survival compared to
th os e pro duc in g fem ale s. Im po r t an tly, the ef fec t s of in cu bat io n tem -
perature persist in sex- reversed individuals, suggesting that temper-
ature, rather than phenotypic sex drives variation in survival. When
coupled with the observation that males reach age at first repro-
duction approximately a decade after females (Zajdel et al., 2019),
our findings offer convincing empirical support for the hypothesis
that differential survival- to- maturity underlies the adaptive value of
TSD. Within this theoretical framework, incubation temperatures
resulting in higher juvenile survivorship are predicted to produce the
sex reaching maturity latest (Schwanz et al., 2016). Previous reports
have shown that TSD species tend to display greater sex biases in
age at mat urity when co mpared to species wi th geno typi c sex deter-
mination (Bókony et al., 2019; Schwanz et al., 2016), yet experimen-
tal studies assessing incubation temperature effects on sur vivorship
within this context are sparse.
It is thought that TSD evolved independently across different
taxonomic groups and likely represents a convergent outcome stem-
ming from different selective pressures (Janzen & Phillips, 2006; Sa rre
et al., 2011; Valenzuela & L ance, 2004). In contrast to the findings
reported here in which incubation temperature affects survivorship,
TSD in jacky dragons likely arose due to incubation temperature-
driven variation in body size and consequent reproductive success
(Warner & Shine, 20 07, 2008). In contrast to many turtle and croc-
odilian species which display 10- to 20- fold longer lifespans, jack y
dragons are among the shor test lived TSD reptiles and la ck substan-
tial sex biases in age at maturity (Warner & Shine, 2008). Given that
most TSD reptiles are longer lived and display stark sex biases in age
at maturity (Bókony et al., 2019), it is possible that incubation tem-
perature affects juvenile survival more broadly and represents the
predominate evolutionary explanation for the adaptive maintenance
of TSD in reptiles. In the common snapping turtle (Chelydra serpen-
tina) hatchlings incubated at a warmer, FPT were shown to have
higher survivorship during their first year in experimental ponds
when compared to hatchlings from a lower, MPT (Janzen, 1995). In
most turtles, including C. serpentina, females typically reach maturity
later than males and the increased survival of female hatchlings ob-
served by Janzen is aligned with the survival- to- maturit y hypothesis
(Christiansen & Burken, 1979). However, in contrast to the results
presented here, the same study showed that turtle hatchlings from
the lower, MPT exhibited increased survival relative to their male
counterparts incubated under intermediate temperatures promot-
ing the development of both sexes (Janzen, 1995). Clearly, additional
studies examining the extent to which incubation temperature influ-
ences juvenile survival under field conditions (Janzen, 1995; Warner
et al., 2020) and the directionality of this relationship with respect to
sex biases in age at maturity are needed to generalize across reptiles
exhibiting TSD.
Incubation temperature exerts clear effects on hatchling mor-
phology, with incubation at MPTs generally resulting in larger and
more massive hatchlings when compared to both cooler and warmer
temperatures. Consistent with our findings, a previous study demon-
strated that incubation at MPTs results in heavier hatchling alligators
with more residual yolk stores, even after considering variation in
egg size (Bock et al., 2021). Developmental cost theory predicts that
the trade- off between temperature- mediated variation in incuba-
tion duration and metabolic rate is optimized at taxon- specific tem-
peratures, at which conversion of maternal resources into offspring
mass is most efficient (Marshall et al., 2020). Our findings suggest
MPTs minimize developmental cost and may contribute to increased
mass and residual yolk stores at hatch by reducing the proportion
of maternally- derived energy reserves used during the incubation
period (Figure 5d; Bock et al., 2021). In contrast, embryos incubated
at FPTs incur a greater cost of development and use a higher pro-
portion of maternal resources to suppor t development leaving less
available in the post- hatching period.
The specific post- hatching traits that mediate the influence of
incubation temperature on early- life survivorship appear more am-
biguous as relationships between hatchling morphology and sur-
vivorship varied across years. Associations between offspring size
and sur vival in reptiles are relatively common in the literature but
few studies have demonstrated direct, causal links between size
and fitness (Janzen et al., 20 07). Further, such associations have
been shown to be context dependent, with size conferring survival
2906
|
Functional Ecology
FUNCTIONAL ECOLOGY
advantages in some years and not others (Ferguson & Fox, 1984;
Olsson & Madsen, 2001; Sinervo et al., 1992 ). Incubation tempera-
ture is commonly reported to affect other hatchling traits (Noble
et al., 2018) including hatchling behaviour (Miller et al., 2020; Nichols
et al., 2019), immune function (Leivesley & Rollinson, 2021; Treidel
et al., 2016), and growth trajectories (Deeming & Ferguson, 1989;
Marcó et al., 2010 ; Piña et al., 2007; Rhen & Lang, 1995), across di-
verse reptile species. It is intriguing to consider that the energetic
cost of development may ser ve as a common underlying mechanism
contributing to multiple fitness- related thermosensitive traits in-
cluding offspring size and metrics of performance via its influence on
the differential allocation of maternally derived resources to devel-
opment versus the post- hatching period. Indeed, emerging evidence
points to developmental cost as a pervasive driver of life history
variation in ectotherms (Marshall et al., 2020; Pet tersen et al., 2020,
2023). However, future studies directly linking developmental en-
ergetics to morphological and physiological traits and subsequent
hatchling survival are required to test this hypothesis. Experimental
approaches incorporating direct manipulation of hatchling energetic
reserves will be particularly informative (Murphy et al., 2020; War-
ner & Lovern, 2014).
The present study does not address how long the influence
of incubation temperature on sur vival persists beyond the first
year of life. However, hatchlings from MPTs displayed enhanced
survival in both pre- winter and post- winter periods, suggesting
that the effect is not limited to one season. Whereas long- term
monitoring of individuals is required to determine the propor-
tion of hatchlings reaching reproductive age, our analysis of re-
ported sex ratios across dif ferent size classes suggests a marked
shift from female- to male- biased sex ratios occurring during the
hatchling- to- juvenile transition. The resulting male bias is then
maintained through the juvenile- to- adult transition, resulting in
broadly obser ved male skews in adult populations. A long- term
mark recapture study in the same alligator population from which
hatchlings in the 2020 and 2021 experiment s originated suggest
apparent survival rapidly increases in juveniles and small adults
relative to hatchlings, but not in a way that differs by sex (Lawson
et al., 2022). When taken together, available evidence suggests
strong influences of incubation temperature on early- life fitness
are likely sufficient to drive differences in sur vival to maturity,
even if these temperature effects wane over time. Our findings
provide empirical support for the hypothesis that differential
FIGURE 5 Temperature- dependent energetic cost of development as a potential mediator of variation in early life survival. (a) Embryonic
metabolic rate (EMR; mL O2/min) measured at stage 26 under the same temperature conditions as those during incubation. (b) Incubation
duration (days) measured as the time between estimated oviposition date and initiation of hatching. (c) Developmental cost measured as
the product of EMR and incubation duration (L O2 consumed during development if metabolic rate were constant). (d) Schematic depicting
hypothesized relationship between incubation temperature, developmental energetic demands and early- life survival. FPT, female
promoting temperature; MPT, male- promoting temperature.
(a)
(d)
(b) (c)
|
2907
Functional Ecology
BOCK et al.
su r v iva l to matu r ity con t rib u tes to th e adap t ive va l ue of TS D in the
American alligator. Future studies which follow the survival and
reproductive outcomes of individuals into adulthood and which
further examine the underlying mechanisms driving temperature-
dependent early- life survival will be critical to unravelling the
adaptive implications of TSD, a long- standing mystery in the field
of evolutionary biology.
AUTHOR CONTRIBUTIONS
Samantha L. Bock and Benjamin B. Parrott conceived the ideas
described in this study; Samantha L. Bock, Yeraldi Loera, David L.
Haskins, Tracey D. Tuberville and Benjamin B. Parrott designed
methodology; Samantha L. Bock, Yeraldi Loera, Josiah M. Johnson,
Christopher R . Smaga, David L. Haskins, Randeep Singh, Thomas R .
Rainwater, Philip M. Wilkinson and Benjamin B. Parrott collected the
data; Samantha L. Bock, Yeraldi Loera, Josiah M. Johnson and Ben-
jamin B. Parrott analysed the data; Samantha L. Bock and Benjamin
B. Parrott led the writing of the manuscript. All authors contributed
critically to the drafts and gave final approval for publication.
ACKNO WLE DGE MENTS
The authors thank Jamie Dozier and the staff at the Tom Yawkey
Wildlife Center for support and assistance in the field. Additional
thanks to numerous students and colleagues that assisted with
hatchling processing and field recaptures, including Marilyn Mason,
Laura Kojima, Kristen Zemaitis, Emily Bertucci- Richter, Ethan Shealy,
Faith Leri, Mark McAlister and Joseph Woods. Thanks also to Maria
Stager for her guidance in the analysis of respirometry data. This
work was partially supported by the Odum School of Ecology and
the National Science Foundation (BBP; award number 1754903).
Furthermore, this mate rial is based upon work supporte d by th e De-
partment of Energy Of fice of Environmental Management (Award #
DE- EM0005228 to the University of Georgia Research Foundation).
This paper represents Technical Contribution Number 7168 of the
Clemson University Experiment Station.
CONFLICT OF INTEREST STATEMENT
The authors declare that there is no conflict of interest.
DATA AVA ILAB ILITY STATE MEN T
Data are available in the Dr yad Digital Repositor y: ht t p s : //d oi .
org/10.5061/dryad.63xsj 3v7c (Bock et al., 2023).
ORCID
Samantha L . Bock https://orcid.org/0000-0002-2124-1490
Yeraldi Loera https://orcid.org/0000-0003-1371-5470
Josiah M. Johnson https://orcid.org/0000-0003-0434-0195
Christopher R. Smaga https://orcid.org/0000-0002-1372-5276
Benjamin B. Parrott https://orcid.org/0000-0002-2391-2470
REFERENCES
Bachtrog, D., Mank, J. E., Peichel, C. L., Kirkpatrick, M., Otto, S. P.,
Ashman, T. L., Hahn, M. W., Kitano, J., Mayrose, I., Ming, R., Perrin,
N., Ross, L ., Valenzuela, N., Vamosi, J. C., Peichel, C. L., Ashman,
T. L., Blackmon, H., Goldberg, E . E., Hahn, M. W., … Vamosi, J. C.
(2014). Sex determination: Why so many ways of doing it? PLoS
Biology, 12(7), 1– 13.
Bates, D., Mächler, M., Bolker, B. M., & Walker, S. C. (2015). Fitting lin-
ear mixed- effect s models using lme4. Journal of Statistical Software,
67(1), 1– 4 8.
Bock, S. L ., Hale, M. D., Rainwater, T. R., Wilkinson, P. M., & Parrot t, B. B.
(2021). Incubation temperature and maternal resource provision-
ing, but not contaminant exposure, shape hatchling phenotypes in a
species with temperature- dependent sex determination. Biological
Bulletin, 241(1), 43– 54.
Bock, S. L., Loera, Y., Johnson, J. M., Smaga, C. R ., Haskins, D. L.,
Tuberville, T. D., Singh, R., Rainwater, T. R., Wilkinson, P. M., &
Parrott, B . B. (2023). Data from: Dif ferential early- life survival un-
derlies the adaptive significance of temperature- dependent sex de-
termination in a long- lived reptile. Dataset. Dryad Digital Repository
https://doi.org/10.5061/dryad.63xsj 3v7c
Bock, S. L., Lower s, R. H., Rainwater, T. R., Stolen, E., Dr ake, J. M., Weiss ,
S., Back, B., Guillette, L. J., & Parrott, B. B. (2020). Spatial and tem-
poral variation in nest temperatures forecasts sex ratio skews in a
crocodilian with environmental sex determination. Proceedings of
the Royal Society B: Biological Sciences, 287, 20200210.
Bock, S. L ., Smaga, C . R., McCoy, J. A ., & Parrott, B. B. (2022). G enome-
wide DNA methylation patterns harbour signatures of hatchling sex
and past incubation temperature in a species with environment al
sex determination. Molecular Ecology, 31(21), 5487– 5505.
Bókony, V., Milne, G., Pipoly, I ., Székely, T., & Liker, A. (2019). Sex ra-
tios and bimaturism differ between temperature- dependent and
genetic sex- determination systems in reptiles. BMC Evolutionary
Biology, 19(1), 1– 7.
Bull, J. J. (1980). Sex determination in reptiles. The Quarterly Review of
Biology, 55(1), 3– 21.
Capel, B. (2017). Vertebrate sex determination: Evolutionary plasticity
of a fundamental switch. Nature Reviews Genetics, 18(11), 675– 689.
Charnier, M. (1966). Ac tion de la temperature sur la sex- ratio chez l'em-
bryon d'Agama agama (Agamidae, Lacertilien). Comptes Rendus D es
Seances de la Societe de Biologie et de Ses Filiales, 160, 620– 622.
Charnov, E. L., & Bull, J. J. (1977). When is sex environmentally deter-
mined? Nature, 266, 828– 832.
Christiansen, J. L., & Burken, R. R. (1979). Growth and maturity of the
snapping turtle (Chelydra serpentina) in Iowa. Herpetologica, 35(3),
261– 266.
Conover, D. O. (1984). Adaptive significance of temperature- dependent
sex determination in a fish. American Naturalist, 123(3), 297– 313.
Conover, D. O., & Kynard, B. E . (1981). Environmental sex determina-
tion: Interaction of temperature and genot ype in a fish. Science,
213(4507), 577– 579.
Deeming, D., & Ferguson, M. (1989). Effects of incubation temperature
on growth and development of embryos of Alligator mississippiensis.
Journal of Comparative Physiology B, 159, 1 8 3 – 193 .
Deitz, D. C. (1979). Behavioral ecology of young American alligators [Ph.D.
disser tation, University of Florida].
Devlin, R. H., & Nagahama, Y. (2002). Sex determination and sex differ-
entiation in fish: An overview of genetic, physiological, and envi-
ronmental influences. Aquaculture, 208(3– 4), 191– 36 4.
Elsey, R. M., & Lang, J. W. (2014). Sex ratios of wild American alligator
hatchlings in Southwest Louisiana. Southeastern Naturalist, 13(2),
191– 199.
Ferguson, G. W., & Fox, S. F. (1984). Annual variation of survival advan-
tage of large juvenile side- blotched lizards, Uta stansburiana: Its
causes and evolutionar y signif icance. Evolution, 38(2), 342– 349.
Ferguson, M. W. J. (1985). Reproduc tive biolog y and embr yolog y of the
crocodilians. In C. Gans, F. Billet, & P. F. A. Maderson (Eds.), Biology
of the Reptilia. Vol. 14. development A (pp. 329– 491). John Wiley &
Sons.
2908
|
Functional Ecology
FUNCTIONAL ECOLOGY
Hartig, F. (2022). Package ‘DHARMa’ residual diagnostics for hierarchic al
(multi- level/mixed) regression models.
Hobaek, A., & Larsson, P. (1990). Sex determination in Daphnia magna.
Ecology, 71(6), 2255– 2268.
Janzen, F. J. (1995). E xperimental evidence for the evolutionary signif-
icance of temperature- dependent sex determination. Evolution,
49(5), 864– 873.
Janzen, F. J., & Phillips, P. C. (2006). Exploring the evolution of envi-
ronmental sex determination, especially in reptiles. Journal of
Evolutionary Biology, 19(6), 1775– 1784.
Janzen, F. J., Tucker, J. K., & Paukstis, G. L . (2007). Experimental analysis
of an early life- history stage: Direct or indirec t selec tion on body
size of hatchling turtles? Functional Ecology, 21, 1 6 2– 1 7 0 .
Katona, G ., Vági, B., Végvári, Z., Liker, A., Freckleton, R. P., Bókony, V.,
& Székely, T. (2021). Are evolutionar y transitions in sexual size
dimorphism related to sex determination in reptiles? Journal of
Evolutionary Biology, 34(4), 594– 603.
Kohno, S., Bernhard, M. C., Katsu, Y., Zhu, J., Bryan, T. A., Doheny, B.
M., Iguchi, T., & Guillette, L. J. (2015). Estrogen receptor 1 (ESR1;
ERα), not ESR2 (ER β), modulates estrogen- induced sex reversal in
the American alligator, a species with temperature- dependent sex
determination. Endocrinology, 156(5), 1887– 1899.
Kohno, S., & G uillet te, L. J. (2013). Endocrine disruption and reptiles:
Using the unique attributes of temperature- dependent sex deter-
mination to assess impac ts. In P. Matthiessen (Ed.), Endocrine dis-
rupters: Hazard testing and asse ssment methods (pp. 245– 271). John
Wiley & Sons.
Korpelainen, H. (1990). Sex ratios and conditions required for envi-
ronment al sex determination in animals . Biological Reviews of the
Cambridge Philosophical Society, 65(2), 147– 184.
Lance, V. A ., Elsey, R. M., & Lang, J. W. (200 0). Sex ratios of American
alligators (Crocodylidae): Male or female biased? Journal of Zoolog y,
252(1), 71– 78.
Lang, J. W., & Andrews, H. V. (1994). Temperature- dependent sex determi-
nation in crocodilians. Journal of Experimental Zoology, 270(1), 28– 44.
Lawson, A. J., Jodice, P. G. R., Rainwater, T. R., Dunham, K . D., Hart, M.,
Butfiloski, J. W., W ilkinson, P. M., McFadden, K. W., & Moore, C.
T. (2022). Hidden in plain sight: Integrated population models to
resolve partially observable latent population structure. Ecosphere,
13(12), e4321.
Leivesley, J. A., & Rollinson, N. (2021). Maternal provisioning and fluc tu-
ating thermal regimes enhance immune re sponse in a reptile with
temperature- dependent sex determination. Journal of Experimental
Biology, 224(5), jeb237016.
Lenth, R. V., Buerkner, P., Giné- Vázquez, I., Herve, M., Jung, M., Love, J.,
Miguez, F., Riebl, H., & Singmann, H. (2023). Package ‘emmeans’:
Estimated marginal means, aka least- squares means (version 1.8.4-
1). American Statis tician, 34(4), 1– 5.
Marcó, P., Piña, C. I., & Simoncini, M. (2010). Effects of incubation and
rearing temperatures on Caiman latirostris growth. Zoological
Studies, 49(3), 367– 373.
Marshall, D. J., Pettersen, A . K ., Bode, M., & White, C. R. (2020).
Developmental cost theory predicts thermal environment and
vulnerability to global warming. Nature Ecology and Evolution, 4(3),
40 6– 411.
Mazerolle, M. J. (2020). Model selection and multimodel inference using
the AICcmodavg package.
Miller, S., Derenne, A., Ellis- felege, S., & R hen, T. (2020). Incubation tem-
perature and satiety influence general locomotor and exploratory
behaviors in the common snapping turtle (Chelydra serpentina).
Physiology & Behavior, 220, 112875.
Murphy, K. M., Radder, R . S., Shine, R., & Warner, D. A. (2020). Lizard
embryos prioritize posthatching energy reser ves over increased
hatchling body size during development. Physiological & Biochemical
Zoology, 93(5), 339– 346.
Nichols, H., Carter, A. W., Paitz, R. T., & Bowden, R. M. (2019). Red- eared
slider hatchlings (Trachemys scripta) show a seasonal shift in behav-
ioral types. Journal of Experimental Zoolog y Part A: Ecological and
Integrative Physiology, 331(9), 485– 493.
Noble, D. W. A ., Stenhouse, V., & Schwanz, L. E. (2018). Developmental
temperatures and phenotypic plasticity in reptiles: A systematic re-
view and meta- analysis. Biological Reviews, 93(1), 72– 97.
Olsson, M., & Madsen, T. (2001). Between- year variation in determinants
of offspring survival in the sand lizard, Lacerta agilis. Functional
Ecology, 15(4), 4 43– 450.
Pettersen, A. K ., Ruuskanen, S., Nord, A., Nilsson, J. F., Miñano, M. R.,
Fitzpatrick, L. J., While, G. M., & Uller, T. (2023). Population diver-
gence in maternal investment and embryo energy use and alloc a-
tion suggests adaptive responses to cool climates. Journal of Animal
Ecology, 1– 15 .
Pettersen, A. K., White, C. R., Bryson- Richardson, R. J., & Marshall, D. J.
(2020). Linking life- history theory and metabolic theory explains
the of fspring size- temperature relationship. Ecology Letters, 22,
518– 526 .
Piña, C. I., Lar riera, A ., Medina, M., & Webb, G. J. W. (2007). Effect s of
incubation temperature on the size of Caiman latirostris (Crocodylia:
Alligatoridae) at hatching and af ter one year. Journal of Herpetology,
41(2), 205– 210.
R Core Team. (2021). R: A language and environment for statistical comput-
ing. R Foundation for Statistical Computing.
Rhen, T., & Lang, J. W. (1995). Phenotypic plasticity for growth in the
common snapping turtle: Effects of incubation temperature,
clutch, and their interaction. The American Naturalist, 146 (5),
72 6 – 747.
Rhodes, W. E., & Lang, J. W. (1996). Alligator nest temperatures and
hatchling sex ratios in coastal South Carolina . In Proceed ings of the
annual conference of the Southeastern A ssociation of Fish an d Wildlife
Agencies ( Vol. 50, pp. 521– 531).
Sabath, N., Itescu, Y., Feldman, A., Meiri, S., Mayrose, I., & Valenzuela,
N. (2016). Sex determination, longevity, and the birth and death of
reptilian species. Ecolog y and Evolution, 6(15), 5207– 5220.
Sæther, B. E., Coulson, T., Grøt an, V., Engen, S., Altwegg, R., Armitage, K.
B., Barbraud, C., Becker, P. H., Blumstein, D. T., Dobson, F. S., Festa-
Bianchet, M., Gaillard, J. M., Jenkins, A ., Jones, C., Nicoll, M. A . C .,
Norris, K., Oli, M. K., Ozgul, A., & Weimer skirch, H. (2013). How
life history influences population dynamics in fluctuating environ-
ments. American Naturalist, 182(6), 743– 759.
Sarre, S. D., Ezaz, T., & Georges, A. (2011). Transitions between sex-
determining systems in reptiles and amphibians. Annual Review of
Genomics and Human Genetics, 12, 3 9 1– 406 .
Schwanz, L. E., Cordero, G. A ., Charnov, E. L ., & Janzen, F. J. (2016). Sex-
specific survival to maturity and the evolution of environmental sex
determination. Evolution, 70(2), 329– 341.
Shine, R. (1999). Why is sex determined by nest temperature in many
reptiles? Trends in Ecology and Evolution, 14 (5), 186– 189.
Sinervo, B., Doughty, P., Huey, R. B., & Zamudio, K. (1992). Allometric
engineering: A causal analysis of natural selection on offspring size.
Science, 258(5 09 0), 1927– 193 0.
Stager, M., Senner, N. R., Swanson, D. L., Carling, M. D., Eddy, D. K.,
Greives, T. J., & Cheviron, Z. A. (2021). Temperature heterogeneity
correlates with intraspecific variation in physiological flexibility in a
small endotherm. Nature Communications, 12(1), 4401.
Subalusky, A. L. (2007). The role of seasonal wetlands in the ecology of the
American alligator [Master of Science, Texas A&M University].
Trei de l, L. A., Ca r te r, A. W., & Bowden, R. M. (2016) . Tem pe ra tu re experi-
enced during incubation affects antioxidant capacity but not oxida-
tive damage in hatchling red- eared slider turtles (Trachemys scripta
elegans). Journal of Experimental Biology, 219(4), 561– 570.
Valenzuela, N., & Lance, V. (Eds.). (20 04). Temperature- dependent sex de-
termination in vertebrates. Smithsonian Books.
|
2909
Functional Ecology
BOCK et al.
Warner, D. A., & Lovern, M. B. (2014). The maternal environment affects
offspring viability via an indirec t effect of yolk investment on off-
spring size. Physiological & Biochemical Zoology, 87 (2), 276– 287.
Warner, D. A., Mitchell, T. S., Bodensteiner, B. L., & Janzen, F. J. (2020).
Sex and incubation temperature independently affect embryonic
development and offspring size in a turtle with temperature-
dependent sex determination. Physiological & Biochemical Zoology,
93(1), 62– 74.
Warner, D. A., & Shine, R. (20 07). Fitness of juvenile lizards depends
on seasonal timing of hatching, not of fspring body size. Oecologia,
154(1), 65– 73.
Warner, D. A., & Shine, R. (2008). The adaptive signific ance of
temperature- dependent sex determination in a reptile. Nature,
451(7178), 566– 568.
White, G. C ., & Burnham, K. P. (1999). Program MARK: Sur vival estima-
tion from populations of mar ked animals. Bird Study, 46, 12 0 – 1 3 9.
Whitehead, P. J., & S eymour, R. S. (1990). Patterns of metabolic rate in
embryonic crocodilians Crocodylus johnstoni and Crocodylus poro-
sus. Physiological Zoology, 63, 33 4– 352.
Wilkinson, P. M., Rainwater, T. R., Woodward, A. R., Leone, E. H., &
Carter, C. (2016). Determinate growth and reproductive lifespan
in the American alligator (Alligator mississippiensis): Evidence from
long- term recaptures. Copeia, 10 4(4), 843– 852.
Zajdel, J., L ance, S. L., Rainwater, T. R., Wilkinson, P. M., Hale, M. D., &
Parrott, B. B. (2019). Mating dynamics and multiple pater nity in a
long- lived vertebrate. Ecology and Evolution, 9(18), 10109– 10121.
SUPPORTING INFORMATION
Additional supporting information can be found online in the
Suppor ting Information section at the end of this article.
Figure S1. I nc ub at io n t e mp er at ur e ef f e ct s o n ha tc hl i ng m o rp h om et r ic
traits— (A) snout- vent length (SVL), (B) body mass, (C) body mass
index (mass/[2 × SVL])— measured shortly after hatching.
Table S1. American alligator sex ratios by size class.
Table S2. Generalized linear mixed effects models of published
alligator population sex ratios.
Table S3. Generalized linear mixed effects models of pre- winter
survival.
Table S4. Generalized linear mixed effects models of post- winter
survival.
Table S5. Li nea r mi xed ef fec ts model s of ha tchl i n g snou t- vent le n g t h .
Table S6. Linear mixed effects models of hatchling mass.
Table S7. Linear mixed effects models of hatchling body mass index.
Table S8. Generalized linear mixed effect s models of pre- winter
survival with hatchling morphological traits as predictors.
Table S9. Generalized linear mixed effects models of post- winter
survival with hatchling morphological traits as predictors.
Table S10. Linear mixed effects models of embryonic metabolic rate
(VO2), incubation duration, and developmental cost.
How to cite this article: Bock, S. L., Loera, Y., Johnson, J. M.,
Smaga, C. R., Haskins, D. L., Tuberville, T. D., Singh, R.,
Rainwater, T. R., Wilkinson, P. M., & Parrott, B. B. (2023).
Differential early- life survival underlies the adaptive
significance of temperature- dependent sex determination in a
long- lived reptile. Functional Ecology, 37, 2895–2909. ht t p s: //
doi .org /10.1111/1365-24 35.14420
