Multiple paternity and breeding system in the gopher tortoise, Gopherus polyphemus.
ABSTRACT Little is known about the reproductive behaviors and the actual outcomes of mating attempts in the gopher tortoise (Gopherus polyphemus). We examined the mating system and reproductive behaviors of a population of gopher tortoises in central Florida. Using microsatellite markers, we assigned fathers to the offspring of seven clutches and determined that multiple fathers were present in two of the seven clutches examined. We found that gopher tortoises exhibited a promiscuous mating system with larger males fertilizing the majority of clutches. The advantage of larger males over smaller males in fertilizing females may be a result of larger males winning access to females in aggressive bouts with other males or larger males may be more attractive to females. Clutches produced by larger females tended to be sired by a single male, whereas clutches of smaller females tended to be sired by multiple males.
- SourceAvailable from: Stephen C. Richter[Show abstract] [Hide abstract]
ABSTRACT: We conducted a genetic study of the largest cluster of US federally threatened Gopher Tortoise (Gopherus polyphemus) colonies. Our objectives were to (1) identify genetic variation within and among colonies across the landscape; (2) determine which factors are important in affecting genetic variation, including land use, habitat quality, and population size; and (3) determine whether genetic partitioning among populations exists and how this relates to (a) geographic distance between sites, (b) Gopher Tortoise natural history and spatial ecology, and (c) land-use history. We studied genetic variability of nine microsatellite DNA loci for 340 adult tortoises from 34 colonies separated by 1.3-45.1 km across a 56,000-ha military installation. Overall genetic variation was low across the landscape and within colonies. Observed heterozygosity (H o ) of tortoise colonies was 49% and allelic richness was 52% of that found in populations located in the eastern portion of the species distribution where habitat is naturally more continuous. Our single colony with highest genetic variation had H o that was 57% and allelic richness that was 60% of eastern colonies. Genetic variation was greatest in sites with suitable habitat. We found weak to no genetic structure across the 45-km landscape (F St = 0.031; Dst = 0.006) and evidence for only one genetic group (K).Although landscape reconfiguration to create sites for military activity has redistributed tortoise colonies and home ranges, we concluded that weak population structure is natural across our study area. Comparison to similar results from a cluster of connected eastern colonies suggests this is a general characteristic of tortoises across large, continuous landscapes and that populations are composed of multiple colonies across the landscape and are naturally large in spatial extent. To alleviate the tortoise-human land use conflict on Camp Shelby, Mississippi, USA and to ensure these created areas continue to benefit tortoises in the long term, maintenance of forest habitat surrounding these created open areas is required. We recommend managing tortoises at Camp Shelby as one unit.Herpetologica 09/2011; 67:406-419. · 1.07 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: In polygynous species, males appear to gain additional offspring by pairing with multiple females simultaneously. However, this may not be true if some females copulate outside of the social pair bond. Polygynous males could experience lower paternity because of trade-offs among gaining multiple social mates, guarding fertility with these mates, and pursuing extra-pair matings. Alternatively, polygynous males could simultaneously gain extra social mates and have high paternity, either because of female preferences or because of male competitive attributes. We tested four predictions stemming from these hypotheses in a facultatively polygynous songbird, the dickcissel (Spiza americana). Unlike most previous studies, we found that males with higher social mating success (harem size) also tended to have higher within-pair paternity and that the number of extra-pair young a male sired increased significantly with his social mating success. Females that paired with mated males were not more likely to produce extra-pair young. In contrast, extra-pair paternity was significantly lower in the nests of females whose nesting activity overlapped that of another female on the same territory. This pattern of mating was robust to differences in breeding density. Indeed, breeding density had no effect on either extra-pair mating or on the association between polygyny and paternity. Finally, nest survival increased with harem size. This result, combined with the positive association between polygyny and paternity, contributed to significantly higher realized reproductive success by polygynous male dickcissels.Behavioral Ecology and Sociobiology 01/2013; 67(2). · 3.05 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: Molecular markers have proven to be a powerful tool for research on turtles. In particular, the application of the polymerase chain reaction (PCR) has increased the availability of molecular technologies while decreasing the cost. However, the cost, time, and expertise associated with developing and testing primers for a particular species can still present a significant barrier, especially to researchers less experienced with molecular methods. In this paper we provide the primer sequence, genomic location, and taxa for 202 PCR primers spanning the entire mitochondrial genome. We also report primers for 11 nuclear coding genes and introns. Finally, we provide primer sequence, amplicon size, and number of observed alleles for 181 microsatellite loci from all major clades of living turtles. We hope that this nearly comprehensive compilation of freshwater turtle and tortoise PCR primers can reduce some of the initial difficulties for beginning turtle geneticists and further facilitate research in existing labs. KEY W ORDS. - Reptilia; Testudines; turtle; PCR; primer; mtDNA; nuclear DNA; microsatellite; STR
Multiple Paternity and Breeding System
in the Gopher Tortoise, Gopherus
JAMIE C. MOON, EARL D. MCCOY, HENRY R. MUSHINSKY, AND STEPHEN A. KARL
From the Department of Biology, University of South Florida, SCA 110, 4202 East Fowler Avenue, Tampa, FL 33620.
Jamie C. Moon is now at the Department of Biology, University of North Florida, 4567 St. John’s Bluff Road South,
Jacksonville, FL 32224. Stephen A. Karl is now at the Hawai’i Institute of Marine Biology, University of Hawaii, Manoa,
PO Box 1346, Kane’ohe, HI 96744.
Address correspondence to S. A. Karl at the address above, or e-mail: firstname.lastname@example.org.
Little is known about the reproductive behaviors and the actual outcomes of mating attempts in the gopher tortoise (Gopherus
polyphemus). We examined the mating system and reproductive behaviorsof a population of gopher tortoises incentral Florida.
Using microsatellite markers, we assigned fathers to the offspring of seven clutches and determined that multiple fathers were
present in two of the seven clutches examined. We found that gopher tortoises exhibited a promiscuous mating system with
larger males fertilizing the majority of clutches. The advantage of larger males over smaller males in fertilizing females may be
be sired by multiple males.
The gopher tortoise(Gopherus polyphemus)isone offour native
North American tortoise species generally occupying dry,
sandy habitats of the southern United States and northern
Mexico. G. polyphemus ranges in the southeast United States
south from the extreme corner of southwestern South Caro-
lina to the tip of the Florida peninsula, west to Louisiana, and
north to about the middle of Mississippi, Alabama, and
Georgia (Auffenberg and Franz 1982). Wild tortoises gener-
ally live for 40–60 years and become reproductively active at
10–20 years of age (Epperson and Heise 2003; Mushinsky
et al. 1994). Although believed to be previously abundant,
the gopher tortoise is declining in numbers, mostly due his-
torically to harvesting and recently to extensive habitat loss
(Auffenberg and Franz 1982; Diemer 1986). Consequently,
the species is of conservation concern and is protected in
a variety of ways throughout its range. G. polyphemus interna-
tionally is listed in Appendix II of the Convention on Inter-
national Trade in Endangered Species of Wild Fauna and
Flora (Inskipp and Gillett 2003); federally is listed by the
sissippi, and west of the Tombigbee and Mobile Rivers in
of Florida is considered a Species of Special Concern by the
Florida Fish and Wildlife Conservation Commission (2005).
Habitat loss in Florida is occurring at an alarming rate.
Between 1936 and 1987, urban land area increased 500%
in Florida (Kautz 1993), and since 1970, the human popula-
tion of Florida has increased by more than 3 million persons
every decade (Bureau of Economic and Business Research
2000). Also between 1936 and 1987, longleaf pine forest,
the prime tortoise habitat, declined in area by 88% (Kautz
1993) with the remaining forests area now small and frag-
servation efforts are needed to ensure the survival of the
In general, sound conservation efforts should include as
much information as possible on the biology and ecology of
a species of interest. In particular, a detailed understanding of
the elements of the mating system can be critical to the per-
sistence of a species. How mates are selected determines in
large part the fate of genetic variation, and genetic variation,
in turn, factors into the presumed long-term evolutionary
1986; Lande and Barrowclough 1987). For turtles in general,
monogamous pair-bond formation is uncommon (Pearse
and Avise 2001). Parental investment is minimal, and the
male’s contribution to the offspring is principally the dona-
tion of sperm. To the contrary, more promiscuous types of
Journal of Heredity 2006:97(2):150–157
Advance Access publication February 17, 2006
ª The American Genetic Association. 2006. All rights reserved.
For permissions, please email: email@example.com.
by guest on June 3, 2013
mating systems, such as polygyny and polyandry, are com-
mon inturtles (Crim etal.2002;Jessopetal.1999;Valenzuela
2000). In a polygynous system, a single male copulates with
multiple females, whereas in a polyandrous system, a single
female copulates with multiple males. There exist, therefore,
differing expectations for patterns of paternity in clutches
produced in polygynous versus polyandrous systems; multi-
ple paternity should be the norm in polyandrous systems but
not necessarily so in polygynous systems. Although female
turtles apparently do not gain direct mating benefits, such
as nuptial gifts or paternal care of offspring, possible indirect
benefits to mating with more than one male would include
gaining ‘‘good genes’’ (Kempenaers et al. 1992; Otter and
Radcliffe 1996; Watson 1998), avoidance of genetic incom-
patibility (Zeh JA and Zeh DW 1996), fertility insurance
(Orsetti and Rutowski 2003), and increased genetic diversity
of offspring (Byrne and Roberts 2000; Madsen et al. 1992).
Many turtle species are known to store sperm for long
tinued to produce viable offspring 4 years after being isolated
from males and thus a source of sperm (Ewing 1943). The
desert tortoise (Gopherus agassizii) has been shown to store
2001). Because of sperm storage, females could engage in
a form of temporal polyandry. Multiple paternity in gopher
tortoises, therefore, could result from inter- as well as intra-
annual copulations. Gopher tortoises exhibit mating behav-
iors from March to September; egg laying, however, is only
known to occur in May and June (Butler and Hull 1996;
Douglass1976; Iverson 1980). Itseems likelythenthatsperm
from late summer or fall mating is stored and used to fertilize
eggs the next spring. Therefore, even females mating with
only a single male within a season, but not between seasons,
would be effectively polyandrous. Clearly, sperm storage
could lead to a higher probability or degree of multiple pa-
ternity even in a strictly monogamous species. Thus, multiple
paternity in gopher tortoises may be an example of temporal
as well as single-season polyandry, the combined effect of
which can be estimated indirectly by assessing the degree
of multiple paternity of natural clutches.
Very little is known about the mating system of gopher
tortoises. Male dominance hierarchies as related to mating
have been documented; larger males most often are the vic-
tors in aggressive male-male interactions (Douglass 1976;
McRae et al. 1981). Douglass (1976) hypothesized that dom-
inant males might also maintain a loose harem (but see
Boglioli et al. 2003). Given the general low density of females
and unfavorable energetics, McRae et al. (1981) suggested
that it would be unlikely for a male to defend several courted
females, and the results of a study by Boglioli et al. (2003)
supported scramble competition polygyny. Instances of male
courtship behaviors (e.g., head bobbing) have been observed
and are commonly reported in wild tortoise populations, al-
though subsequent copulation rarely has been observed ex-
cept in captive settings (Auffenburg 1966; Douglass 1976;
Wright 1982). Male-male confrontations, in which the partic-
ipants attempt to overturn one another, also are observed
during courting season (Hailman 1991; personal observa-
tion). The importance of these behaviors, however, is largely
unknown. Field observation can identify dominant males on
an incident-by-incident basis, but rarely can they reveal
whether dominance leads to greater reproductive success.
in a clutch is necessary. Molecular genetic techniques able to
identify single individuals are providing answers to a variety
of questions concerning mating systems, such as paternity
(Roques et al. 2004), maternity (Jones et al. 1999), and mating
behaviors (Strausberger and Ashley 2003). Microsatellite loci
are by far the most frequently used genetic tool in the inves-
tigation of specifics of mating systems. These loci mutate
quickly and, therefore, are highly polymorphic (Dallas 1992)
making them especially useful in identifying individuals
(Queller et al. 1993). The probability of identifying or exclud-
ing from consideration specific individuals often is extremely
large with multilocus microsatellite genotypes. Given this,
microsatellite markers are an ideal genetic tool for assaying
Here, we report the results of a microsatellite locus geno-
typing study examining the mating system of a population
of gopher tortoises in central Florida. In particular, we ad-
dressed the following questions: Does multiple paternity oc-
cur in gopher tortoise clutches, and, if so, is the fertilization
of a single clutch evenly divided among the males? Do some
males fertilize more clutches than others, and, if so, is fertil-
ization success related to male size?
Our field site was the University of South Florida Ecological
Research Area (ERA), a 200-ha reserve located in Hillsbor-
proximately 20 ha of sandhill habitat within the ERA have
been exposed to controlled burning since 1976 (Mushinsky
1992). This area is separated into 11 plots that are burned
on 1-, 2-, 5-, or 7-year cycles or left as unburned controls
(Mushinsky 1985). Tortoises are distributed through the up-
land habitat of the ERA but are most commonly found in the
plots subjected to control burns.
A population of about 280 tortoises occupies the ERA
(Mushinsky et al. 1994). All plots were trapped for tortoises
during the course of this study from April to August of 2001
and again in 2002, and all active and inactive burrows (based
on Mushinsky and McCoy 1994) in each plot were located
and marked. The width of each burrow was measured at
a depth of 500 mm and used as an estimate of the carapace
length (CL) of the resident tortoise (Wilson et al. 1991). Bur-
rows greater in width than the minimum CL of a sexually
mature individual of eithersex (i.e., males at 177 mm;Diemer
and Moore 1994; Mushinsky et al. 1994) were pit trapped by
burying at the burrow entrance a 9.5-L bucket camouflaged
with brown fabric and sand with the opening level with the
surface of the ground. While set, the traps were checked ev-
ery 2 h during daytime. Individuals also were captured by
hand when encountered outside their burrows.
Moon et al.?Multiple Paternity in Gopher Tortoise
by guest on June 3, 2013
Sex and stage of maturity in gopher tortoises can be
inferred readily by the size of the carapace. Tortoises larger
than 240-mm CL display sex-specific degrees of plastral con-
cavity (PC). The PC of females is less, and that of males
greater, than 6 mm deep (Mushinsky et al. 1994). Females
in the ERA are reproductive at ;240-mm CL (Mushinsky
et al. 1994) and males at CL from 177 to 230 mm (Diemer
and Moore 1994).
Mass and CL were assessed for all sexually mature and
subadult tortoises captured (Mushinsky et al. 1994). Differ-
ences between various groupings of individuals were tested
using Student’s t tests. The CL and mass of the mothers and
the fathers assigned to the clutch based on the genetic data
for single- and multiple-sire clutches, as well as clutch char-
acteristics, also were compared using Student’s t tests. A
blood sample was obtained from the brachial sinus with
a heparinized 21-gauge needle and a 10-cc syringe. After
collection, blood was stored in a lysis buffer (10 mM Tris-
HCl pH 8.0, 100 mM NaCl, 50 mM ethylenediaminetetra-
acetic acid, 1% w/v polyvinylpyrrolidone, and 0.2% v/v
2-mercaptoethanol) until DNA extraction. Unless already
by notching the marginal scutes, and all individuals were re-
leased at the site of capture. The overall goal was to genotype
all sexually mature males in the population (i.e., the paternal
pool) and genotype and capture all sexually mature females.
To determine if they were carrying shelled eggs, sexually
mature females were taken to the laboratory and radiogra-
phed with one pulse from The Inspector X-ray Source,
Model 200 (Golden Engineering, Centerville, Indiana), with
a source output of 3 mR at 1 ft per pulse. Females with com-
pletely shelled eggs were given an injection of 1.5 units per
100-g body mass of 3% oxytocin to stimulate oviposition
(Ewert and Legler 1978; Iverson J, personal communication).
After oviposition, blood was drawn and stored for DNA
extractions, and mothers were returned to the site of capture.
Eggs were incubated in the laboratory in nests of vermic-
ulite with a 1:1 w/v ratio of vermiculite to water (Burke et al.
1996) and the temperature maintained at ;30?C (Burke et al.
1996; Demuth 2001). Eggs were inspected daily during incu-
bation and hatchlings removed when discovered. Eggs that
did not hatch after 120 days were removed and opened to
remove the embryo, if present, and frozen for later DNA
extraction. Live hatchlings were maintained in aquaria with
a 12:12 light:dark cycle and fed a diet of vegetables supple-
mented with vitamin and mineral powders. Blood samples
were obtained from hatchlings by cardiocentesis with a hep-
arinized 26-gauge needleand a1-cc syringe (Jacobson E,per-
sonal communication). Hatchlings were given individual
markings in the same manner as adults and released at the
burrow where the mother was captured.
Total cell DNA was extracted following a standard phenol/
chloroform protocol and ethanol precipitated (Herrmann
and Frischauf 1987). All individuals were genotyped at nine
microsatellite loci (GP15, GP19, GP26, GP30, GP55,
GP61, GP81, GP96, and GP102) previously characterized
by Schwartz et al. (2003). Microsatellite polymerase chain
eters following Schwartz et al. (2003). After amplification,
samples were run on an ABI Prism 377 automated DNA se-
quencer (Perkin Elmer, Applied Biosystems, Inc, Foster City,
CA) at Iowa State University. The program, GENESCAN
ensure accuracy in reading and recording of results.
Maternalcontribution ofeach individual intheclutch was
determined by comparing the offspring’s and the mother’s
genotypes. Paternal alleles were inferred by removing known
maternal allelic contributions from the offspring genotypes,
when possible. If among all hatchlings, more than two pater-
nal alleles were found at any locus, the clutch was classifiedas
having multiple sires. For multiply sired clutches, paternity
exclusion was done on an individual hatchling basis. In any
case, candidate males whose genotype mismatched the pater-
nal contribution of the hatchlings at one or more loci were
excluded as potential sires. Individual and combined exclu-
CERVUS,version 2.0 (Marshalletal.1998). Whenmorethan
performed a probability analysis on all nonexcluded males.
CERVUS uses a log (base e) likelihood algorithm to calculate
the likelihood ratio (LOD) of a candidate male being the true
parent compared to an arbitrary male. LOD scores are calcu-
lated for all possible sires (i.e., nonexcluded males), and the
difference between the two most likely candidates (D) is cal-
culated and provides an indication of the reliability of the as-
signment. By simulation, a critical D score is calculated at
either relaxed (80%) or strict (95%) confidence level and cor-
responds to an estimated frequency of false positives of 20%
and 5%, respectively. In multiple sire clutches, it is possible
that all males except one are excluded as potential fathers
for some, but not all, of the hatchlings. When the unequivo-
cally assigned father was one of the nonexcluded fathers for
a clutch mate with more than one potential sire, he was
assigned as the father regardless of the D value.
Allele frequencies were estimated only from wild-caught
individuals (i.e., not including hatchlings). The probability of
(probability of identity) was calculated as in Hanotte et al.
(1991). The probability of exclusion (D) is the overall prob-
ability of detecting multiple paternity and was calculated as in
Westneat et al. (1987). This calculation utilizes the frequency
of alleles and all possible mating arrangements to compute
the probability of detecting when a female has mated with
We captured 15 adult females representing approximately
31% of the 49 adult females observed in the ERA between
1988 and 1993. Seven of the 15 were gravid and produced
Journal of Heredity 2006:97(2)
by guest on June 3, 2013
clutches with the number of offspring per clutch ranging
from 4 to 11 with a mean of 7.57 ± 2.44 (Table 1). A total
of 90 individuals were genotyped, including 5 subadults, 15
adult females, 17 adult males, and 53 offspring. The nine
microsatellite loci had two to five alleles each, with a mean
of 3.44 ± 1.01 alleles per locus and observed per-locus het-
erozygosity ranged from 0.053 to 0.658 (Table 2). The aver-
age probability of two unrelated tortoises sharing the same
genotype at all nine loci (probability of identity) was 3.84 ?
10?4(Table 2). The probability of detecting multiple pater-
nity (D) with a single locus ranged from 0.026 (at GP96) to
0.410 (GP81; Table 2) and was 0.876 when all nine loci
were used. The cutoff D value estimates for the relaxed and
strict confidence limits were D80%5 1.44 and D95%5 3.10,
In most cases, the maternal contribution could be unambig-
uously determined by inspection. In cases where the mother
and offspring were both heterozygous for the same alleles,
determining unambiguously which allele was the maternal al-
lele was impossible. Males who possessedat leastone copy of
either allele, therefore, could not be excluded from the anal-
ysis. When a father was assigned to a clutch or an individ-
ual, the probability of another unrelated tortoise having
the same genotype as the father was extremely low, ranging
from 5.65 ? 10?13to 1.37 ? 10?6(Table 3).
The genetic contribution of the male also was often un-
ambiguously identified. In four of the seven clutches, all can-
didate males but one were excluded because, at one or more
of the loci, they could not have contributed the paternal allele
to the offspring (Table 3). For one hatchling in a fifth clutch
(378), all but one male (519) were excluded as potential sires.
For each of the remaining three clutch mates, however, three
candidate males could not be excluded based on genotype
alone. Results from an analysis with CERVUS indicate that
male 519 was the most likely father to all three with at least
80% confidence (2.52 ? D ? 3.20), so we assigned this male
as thesole father ofthe clutch. In theremainingtwo clutches,
three or more paternal alleles were found at one or more loci,
indicating multiple sires for these clutches. For one hatchling
in one of these clutches (446), all males except 533 could be
excluded as a sire. For the remaining clutch mates, however,
male 43 was either the only nonexcluded male (two individ-
uals) or the most likely sire in the likelihood analysis (two
individuals) albeit with small D values (0.056 ? D ?
0.064). Male 43 also was the sole putative sire for one of
the single-sire clutches (Table 3). In the other multiple-sire
clutch (523), males 180 and 265 were each either the only
nonexcluded male or one of several nonexcluded fathers
(Table 3). For one individual in this clutch, all the males in-
cluded in this survey were excluded based on genotype alone
because this hatchling had one allele at each of two loci not
seen in any male surveyed.
Reproductive output and hatching success were variable
among females. The number of eggs in the clutch did not
vary significantly between single- and multiple-sire clutches
(tdf555 0.703, P 5 .513; Table 1). The number of undevel-
oped eggs also did not vary between single- and multiple-sire
Characteristics of mother, clutch, and father for single-sired and multiple-sired clutches
Single siredMultiple sired
Clutch 18Clutch 201Clutch 253Clutch 378 Clutch 529 Clutch 446 Clutch 523
Mother CL (mm)
Mother mass (g)
Father CL (mm)
Father mass (g)
Total number of eggs
Hatching success (%)
Number of undeveloped eggs
Number of eggs genotyped
% Fertilized by 1? male
aThis male sired the largest number of the clutch mates and is designated as the primary male.
bA blood sample could not be obtained from one member of this clutch.
(HE) heterozygosity, probability of identity, and probability of
exclusion for nine microsatellites in Gopherus polyphemus
The number of alleles, observed (HO) and expected
3.44 ± 1.01
3.84 ? 10?4
Moon et al.?Multiple Paternity in Gopher Tortoise
by guest on June 3, 2013
clutches (tdf555 ?1.641, P 5 .162; Table 1). A negative
relationship exists between hatching success and multiple
paternity (tdf555 2.254, P 5 .074).
Some morphological differences were noted but were
limited to differences among females of single- and multi-
ple-sired clutches. Data on mass and CL were normally dis-
tributed and homoscedastic. Males who were assigned as
fathers (N 5 8) had larger CL than unassigned males
(N 5 9; tdf5155 2.400, P 5 .030) but did not differ in mass
(tdf 5 155 1.335, P 5 .202). Mass (tdf565 ?0.967, P 5 .371)
and CL (tdf565 ?0.053, P 5 .960) were not significantly
different between fathers of single-sire (N 5 5) and multi-
ple-sire clutches (N 5 4). Mass (tdf555 6.409, P 5 .001)
and CL (tdf555 2.682, P 5 .044) did differ between females
inseminated by a single male (N 5 5) and females insemi-
nated by more than one male (N 5 2), however. Females
inseminated by multiple males on average weighed 1320 g
less and had CLs 23.1 mm shorter than females inseminated
by a single male (Table 1).
All mothers were genotyped, and no mismatches between
known mother-offspring pairs were observed. The mother’s
contribution to each clutch was identified and removed, and
all remaining alleles were designated as paternal alleles. Mul-
tiple paternity was observed in two of the seven clutches. In
general, the percentage of clutches in which multiple pater-
nity is found varies with species, and sometimes even among
populations within a species. For those turtle species that
have been studied, the percentage of multiply fertilized
clutches ranges from 4% to 100% (Hoekert et al. 2002;
Moore and Ball 2002; Pearse and Avise 2001). In this study,
28.6% showed multiple paternity. This number may be con-
servative, however, as the ability of the loci used in this study
to detect multiple paternity was only 87.6%. Thus, for each
clutch in which one father was detected, there was a 12.4%
chance that another father went undetected. A full character-
ization of the frequency of multiple paternity in gopher tor-
toises will require a much more extensive survey including
several populations from throughout the range and more
individuals per population.
One offspring, in multiple sire clutch 529, could not be
assigned to any of the genotyped males. One possibility is
that this individual was indeed the offspring of one of the
surveyed males but appears different from either the mother
or father as a result of new mutation. Because there are two
happenedindependently at each of them concurrently, which
is unlikely. A second, more reasonable, possibility is that
three different males, one of which was not among the geno-
typed males, inseminated the female. Between 1988 and
1993, 42 adult males have been identified in the area, but
we genotyped only 17 (41%). A third possibility is that
the female may be a recent immigrant into the area. Al-
though all individuals in this area have been routinely marked
(Mushinsky et al. 1994), the mother of the unassignable off-
spring was not. It is possible then that in addition to the
sperm from the two local males the female was carrying
sperm from at least one male outside the ERA. Given the
possibility of long-term sperm storage in gopher tortoises,
this latter scenario is possible. Regardless, multiple paternity
is clearly observed in gopher tortoises and may constitute
a frequent mating outcome. We cannot, however, differen-
tiate between intraseasonal and temporal polyandry.
Reports of aggressive interactions between male gopher
tortoises have provoked the question of whether or not these
Results of paternity assignment and probability of identity for the seven clutches and females surveyed in this study
Putative single sire
1.14 ? 10?10
4.11 ? 10?13
5.65 ? 10?13
2.82 ? 10?8
1.37 ? 10?6
Putative multiple sire
7.07 ? 10?12
1.14 ? 10?10
5.44 ? 10?8
1.58 ? 10?8
aCalculated using CERVUS when there was more than one nonexcluded male and is the range over all offspring for the assigned male. Critical delta values for
the relaxed and strict confidence levels are 1.44 and 3.10, respectively.
bThis is the total number of unique fathers over all of these offspring.
Journal of Heredity 2006:97(2)
by guest on June 3, 2013
duels influence a male’s access to females. If winning males
do receive greater access to females, then the victor’s fitness
would likely be higher. If the chances of winning are posi-
more females and fertilize more eggs. In this study, one large
male (43) fertilized two of the seven gravid females sampled
(nearly 30%) and 12 of the 53 eggs (23%). The remaining
eight fathers each fertilized only one female and from
1.2% to 20.7% of the eggs. Males that were assigned as sires
of the clutches were significantly larger than the unassigned
males, suggesting that larger males have a reproductive ad-
vantage. Larger males often have been reported as the win-
observed aggressive male-male interactions during the breed-
ing season near the entrances to female burrows. Thus, it is
likely that the greater proportion of fertilization attributable
to larger males is at least partially a result of winning aggres-
sive encounters and gaining greater access to females.
Adult size also may be a significant factor regulating mul-
ferent between single- and multiple-sired clutches. Females
with multiple-sired clutches were lighter and shorter than
females with single-sired clutches (Table 1). At least two
explanations exist for this difference. The first involves male
guarding behavior. Male-male duels and male-female court-
ships in gopher tortoises can take a considerable amount of
time (Douglass 1976; McRae et al. 1981). Even males that are
generally successful in aggressive encounters, therefore, may
be temporally limited in the number of females with which
they can mate. If males are limitedto only a few females, then
malesmayprefer,and preferentiallyguard, largerfemales that
can produce moreeggs (Pearse et al. 2002). If larger males are
more likely to select and successfully guard large females,
then large males should on average sire clutches from large
females. The correlation (R 5 .852, P 5 .067) of female ver-
sus male parent mass for single-sired clutches is consistent
with thisexpectation (data availableon request).Larger males
preferentially guarding larger females also would result in
smaller females either being unguarded or ineffectively
guarded by small males. It seems possible then that this could
result in smaller females being more frequently inseminated
by multiple males. After having copulated with a female,
a small male could be displaced by a larger male, which,
depending on the female’s ability to fend off the male,
may result in a multiple-sired clutch. Although our sample
size is somewhat small, our results are consistent with this
in that the two females producing multiple-sired clutches
were lighter and had shorter CL than all females with
single-sired clutches (Table 1).
A second explanation for differences in female size for
single- and multiple-sired clutches involves female receptive-
ness. In the field, females have been observed blocking their
burrow entrances and rebuffing males performing courting
behaviors (Douglass 1976; personal observation). Larger
females may be better able to turn away attempts by courting
males than are smaller females, especially when the courting
males are large. If this were the case, then, again, we would
expect the incidences of multiple paternity to be higher in
small females, as is observed. It is important to note that
we cannot differentiate between explanations involving size
versus age (i.e., experience) because age and size are strongly
correlated. Smaller and lighter females are likely also to rep-
resent younger and less experienced females. Novice females
may be engaging in different behaviors simply because they
are inexperienced, and any difference may reflect experience
and not size, per se. Younger females may also be behavior-
ally as well as physically less selective in choosing mates.
Regardless of the proximate cause of multiple-sired
clutches, a clear result is that promiscuity, both in the form
of polyandry and polygyny, is occurring in gopher tortoises.
From a conservation standpoint, this is the most positive sit-
uation. Increases in the variance of reproductive success (i.e.,
polygyny) reduce the genetically effective population size
(Ne) relative to the census size (Wright 1938). Reductions
in Necan lead to inbreeding and higher probabilities of
expressing deleterious recessive alleles resulting in reduced
fitness. Even so, the addition of polyandry and a long gen-
eration time (i.e., overlapping generations) likely are amelio-
rating factors (Nunney 1993) partially, if not completely,
compensating for any reductions in Ne. With these initial
data, however, we cannot fully determine the evolutionary
or conservation implications or how typical the ERA popu-
lation is of other tortoise populations. However, it does ap-
pear that the reproductive output and probability of mating
with more than one male are related to female size or expe-
rience (at least to the extent that they are correlated). Further
ulation are necessary to fully characterize the consequences
of the promiscuous mating system in gopher tortoises.
This project was supported by a J. Landers Student Research Award to
J.C.M. and Arcadia Wildlife Preserve, Inc. and National Science Foundation
Grant in Systematics DEB 98-06905 to S.A.K. We thank Tonia Schwartz for
her aid with laboratory procedures and three anonymous reviewers for
helpful comments on an earlier draft. Animals were trapped and collected
under Florida Fish and Wildlife Service permit WV01274, and tissue collec-
tion and handling procedures were approved by The University of South
Florida IACUC permit number 1742.
Auffenberg W, 1966. On the courtship of Gopherus polyphemus. Herpetologica
Auffenberg W and Franz R, 1982. The status and distribution of the gopher
tortoise (Gopherus polyphemus). In: North American tortoises: conservation
and ecology (Bury RB, ed). Washington, DC: U.S. Department of Interior,
Fish and Wildlife Service, Wildlife Research Report 12; 95–126.
Boglioli MD, Guyer C, and Michener WK, 2003. Mating opportunities of
females gopher tortoises, Gopherus polyphemus, in relation to spatial isolation
of females and their burrows. Copeia 2003:846–850.
Bureau of Economic and Business Research, 2000. Florida statistical Ab-
stract. Gainesville, FL: University Presses of Florida.
Burke RL, Ewert MA, McLemore JB, and Jackson DR, 1996. Temperature-
dependent sex determination and hatching success in the gopher tortoise
(Gopherus polyphemus). Chelonian Conserv Biol 2:86–88.
Moon et al.?Multiple Paternity in Gopher Tortoise
by guest on June 3, 2013
Butler JA and Hull TW, 1996. Reproduction of the tortoise, Gopherus poly-
phemus, in northeastern Florida. J Herpetol 30:14–18.
Byrne PGandRobertsJD, 2000. Does multiplepaternityimprovethe fitness
of the frog Crinia georgiana? Evolution 54:968–973.
Crim JL, Spotila D, Spotila R, O’Connor M, Reina R, and Williams CJ, 2002.
The leatherback turtle Dermochelys coriacea, exhibits both polyandry and po-
lygyny. Mol Ecol 11:2097–2106.
Dallas JF, 1992. Estimation of microsatellite mutation rates in recombinant
inbred strains of mouse. Mamm Genome 5:32–38.
Demuth JP, 2001. The effects of constant and fluctuating incubation tem-
perature on sex determination, growth, and performance in the tortoise
Gopherus polyphemus. Can J Zool 79:1609–1620.
Diemer JE, 1986. The ecology and management of the gopher tortoise in the
southeastern United States. Herpetologica 42:125–133.
north-central Florida.In: Biology of North American tortoises (Bury RB and
Wildlife Research Report 13; 129–137.
Douglass JF, 1976. The mating system of the gopher tortoise, Gopherus poly-
phemus, in southern Florida (MS thesis). Tampa, FL: University of South
Dudash M and Fenster CB, 2000. Inbreeding and outbreeding depression in
fragmented populations. In: Genetics, demography and viability of frag-
mented populations (Young AG and Clarke GM, eds). New York: Cam-
bridge University Press; 35–53.
tortoises (Gopherus polyphemus) in southern Mississippi. J Herpetol 37:315–
Ewert MA and Legler JM, 1978. Hormonal induction of oviposition in tur-
tles. Herpetologica 34:314–318.
Ewing HE, 1943. Continued fertility in female box turtles following mating.
Florida Fish and Wildlife Conservation Commission, 2005. Florida’s endan-
gered species, threatened species, and species of special concern. Available
at: http://www.wildflorida.org. Accessed April 27, 2005.
Gilpin ME and Soule ´ ME, 1986. Minimum viable populations; processes of
species extinction. In: Conservation biology (Soule ´ ME, ed). Sunderland,
MA: Sinauer Associates, Inc.; 19–34.
Hailman JP, 1991. Notes on the aggressive behavior of the gopher tortoise.
Herpetol Rev 22:87–88.
isatellite DNA sequences in the Indian peafowl Pavo cristatus. Genomics
Herrmann BG and Frischauf A, 1987. Isolation of genomic DNA. Methods
Hoekert WEJ, Neufe ´glise H, SchoutenAD, and Menken SBJ, 2002. Multiple
paternity and female-biased mutation at a microsatellite locus in the olive
ridley sea turtle (Lepidochelys olivacea). Heredity 89:107–113.
Inskipp T and Gillett HJ, 2003. Checklist of CITES species. Compiled by
UNEP-WCMC, CITES Secretariat, Geneva, Switzerland and UNEP-
WCMC, Cambridge, UK; 339 pp. and CD-ROM.
Iverson JB, 1980. The reproductive biology of Gopherus polyphemus (Chelonia:
Testudinidae). Am Midl Nat 103:353–359.
Jessop TS, FitzSimmons NN, Limpus CJ, and Whittier JM, 1999. Interac-
tions between behavior and plasma steroids within the scramble mating sys-
tem of the promiscuous green turtle, Chelonia mydas. Horm Behav 36:
Jones AG, Rosenqvist G, Berglund A, and Avise JC, 1999. The genetic mat-
ing system of a sex-role-reversed pipefish (Syngnathus typhle): a molecular in-
quiry. Behav Ecol Sociobiol 46:357–365.
Kautz RS, 1993. Trends in Florida wildlife habitat 1939–1987. Fla Sci 56:
Kempenaers B, Verheyen GR, Vandenbroeck M, Burke T, Vanbroeckhoven
for high-quality males in the blue tit. Nature 357:494–496.
Lande R and Barrowclough GF, 1987. Effective population size, genetic var-
iation, and their use in population management. In: Viable populations for
conservation (Soule ´ ME, ed). Cambridge, MA: Cambridge University Press;
MadsenT,ShineR,LomanJ,andHakansson T,1992.Why dofemale adders
copulate so frequently? Nature 355:440–441.
Marshall TC, Slate J, Kruuk LEB, and Pemberton JM, 1998. Statistical con-
fidence for likelihood-based paternity inference in natural populations. Mol
range of the gopher tortoise. Am Midl Nat 106:165–179.
Moore MK and Ball RM Jr, 2002. Multiple paternity in loggerhead turtle
(Caretta caretta) nests on Melbourne Beach, Florida: a microsatellite analysis.
Mol Ecol 11:281–288.
Mushinsky HR, 1985. Fire and the Florida sandhill herpetofaunal com-
munity: with special attention to responses of Cnemidophorus sexlineatus.
Mushinsky HR, 1992. Natural history and abundance of southeastern five-
lined skinks, Eumeces inexpectatus, on a periodically burned sandhill in Florida.
Mushinsky HR and McCoy ED, 1994. Comparison of gopher tortoise pop-
ulations on islands and on the mainland in Florida. In: Biology of North
American tortoises (Bury RB and Germano DJ, eds). Washington, DC:
U.S. Fish and Wildlife Service, Fish and Wildlife Research Report 13:
Mushinsky HR, Wilson DS, and McCoy ED, 1994. Growth and sexual
dimorphism of Gopherus polyphemus in central Florida. Herpetologica 50:
Noss RF, 1989. Longleaf pine and wiregrass: keystone components of an
endangered ecosystem. Nat Areas J 9:211–213.
Nunney L, 1993. The influence of mating system and overlapping genera-
tions on effective population size. Evolution 47:1329–1341.
Orsetti DM and Rutowski RL, 2003. No material benefits, and a fertilization
cost, for multiple mating by female leaf beetles. Anim Behav 66:477–484.
Otter K and Ratcliffe L, 1996. Female initiated divorce in a monogamous
songbird: abandoning mates for males of higher quality. Proc R Soc Lond
PalmerKS, RostalDC,GrumblesJS, andMulvey M, 1998. Long-termsperm
storage in the desert tortoise (Gopherus agassizii). Copeia 1998:702–705.
Pearse DE and Avise JC, 2001. Turtle mating systems: behavior, sperm stor-
age, and genetic paternity. J Hered 92:206–211.
Pearse DE, Janzen FJ, and Avise JC, 2002. Multiple paternity, sperm storage,
and reproductive success of female and male painted turtles (Chrysemys picta)
in nature. Behav Ecol Sociobiol 51:164–171.
Queller DC, Strassmann JE, and Hughes CR, 1993. Microsatellites and kin-
ship. Trends Ecol Evol 8:285–288.
Roques S, Diaz-Paniagua C, and Andreu AC, 2004. Microsatellite markers
tortoise, Testudo graeca. Can J Zool 82:153–159.
Schwartz TS, Osentoski M, Lamb T, and Karl SA, 2003. Microsatellite loci
for the North American tortoises (genus Gopherus) and their applicability to
other turtle species. Mol Ecol Primer 3:283–286.
Strausberger BM and Ashley MV, 2003. Breeding biology of brood parasitic
brown-headed cowbirds (Molothrus ater) characterized by parent-offspring
and sibling-group reconstruction. Auk 120:433–445.
Journal of Heredity 2006:97(2)
by guest on June 3, 2013
U.S. Fish andWildlifeService, 1986. Endangeredandthreatened wildlife and
plants: determination of threatened status for the gopher tortoise (Gopherus
polyphemus). Fed Regist 52:25376–25380.
Valenzuela N, 2000. Multiple paternity in side-neck turtle Podocnemis expansa:
evidence from microsatellite DNA data. Mol Biol 9:99–105.
Watson PJ, 1998. Multi-male mating and female choice increase offspring
Westneat DF, Fredrick PC, and Wiley RH, 1987. The use of genetic markers
to estimate the frequency of successful alternative reproductive tactics.
Behav Ecol Sociobiol 21:35–45.
Wilson DS, Mushinsky HR, and McCoy ED, 1991. Relationship between
gopher tortoise body size and burrow width. Herpetol Rev 22:122–124.
Wright JS, 1982. The distribution and population biology of the gopher
tortoise (Gopherus polyphemus) in South Carolina (MS thesis). Clemson, SC:
Wright S, 1938. Size of population and breeding structure in relation to evo-
lution. Science 87:430–431.
Zeh JA and Zeh DW, 1996. The evolution of polyandry I: intragenomic con-
flict and genetic incompatibility. Proc R Soc Lond B 263:1711–1717.
Received July 22, 2005
Accepted November 30, 2005
Corresponding Editor: Brian Bowen
Moon et al.?Multiple Paternity in Gopher Tortoise
by guest on June 3, 2013