This article appeared in a journal published by Elsevier. The... copy is furnished to the author for internal non-commercial research

This article appeared in a journal published by Elsevier. The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies are
encouraged to visit:
http://www.elsevier.com/copyright
Author's personal copy
Animal Behaviour 78 (2009) 747–753
Contents lists available at ScienceDirect
Animal Behaviour
journal homepage: www.elsevier.com/locate/yanbe
Potential reproductive rate of a sex-role reversed pipefish over
several bouts of mating
Sunny K. Scobell*, Adam M. Fudickar 1, Rosemary Knapp 2
Department of Zoology, University of Oklahoma
a r t i c l e i n f o
Article history:
Received 5 February 2009
Initial acceptance 19 March 2009
Final acceptance 28 May 2009
Published online 4 August 2009
MS. number: A09-00073
Keywords:
gulf pipefish
life-history theory
ovarian morphology
pipefish
potential reproductive rate
sex-role reversal
sexual selection
Syngnathidae
Syngnathus scovelli
The potential reproductive rate (PRR, the rate at which each sex could reproduce if given unlimited
mates) has proven to be a useful tool in predicting the direction and strength of sexual selection. We
conducted a 2-month study of the PRR in the polyandrous gulf pipefish, Syngnathus scovelli, a year-round
breeder. In this sex-role reversed species, the female transfers eggs to a male’s brood pouch during
mating and thus renders him unavailable to mate for 2 weeks. We predicted females would have a higher
PRR than males and that the rates in both sexes would change over successive breeding bouts in relationship to previous reproductive output. Females did have a higher overall PRR than males for the entire
study period. However, PRR was not constant across individual breeding bouts. For each sex, the PRRs
from the first and third bouts of mating were significantly higher than the PRR of second mating bout.
Our results are consistent with individuals making trade-offs between current and future reproductive
investment. We also discuss how ovarian morphology may contribute to elevated female PRR in this
species. To our knowledge, this is the first study of PRR in a North American pipefish.
The Association for the Study of Animal Behaviour. Published by Elsevier Ltd.
When studying mating behaviour, researchers usually want to
know which sex is under greater sexual selection pressure and to
what degree. This information is useful in determining the sex that
should be competitive for mates and the sex that should be choosy.
The strength of sexual selection acting on each sex can also be
correlated with territoriality, the degree of sexual dimorphism and/
or ornamentation and variance in mating success (Andersson
1994). However, measuring the strength of sexual selection acting
on each sex has been difficult historically, and determining which
methodology is best has been the subject of debate.
Many researchers have measured sex differences in parental
investment and/or the operational sex ratio, two indirect measures
of the strength of sexual selection. Trivers (1972) proposed that the
sex that has relatively greater parental investment (PI) should be
choosy, leaving the sex that has lower PI to compete for mates.
* Correspondence and present address: S. K. Scobell, Department of Biology,
Texas A&M University, 3258 TAMU, College Station, TX 77843-3258, U.S.A.
E-mail address: [email protected] (S.K. Scobell).
1
A. M. Fudickar is now at the Department of Migration and Immuno-ecology,
Max Planck Institute for Ornithology, Radolfzell, Germany.
2
R. Knapp is at the Department of Zoology, University of Oklahoma, 730 Van
Vleet Oval, Room 314, Norman, OK 73019, U.S.A.
However, PI (defined as any effort that increases offspring survival
at the expense of the parent’s ability to invest in other offspring) can
be difficult to measure, particularly in species where both sexes
have some form of investment (Clutton-Brock 1991). Emlen & Oring
(1977) proposed that a bias in the operational sex ratio (OSR) away
from 1:1 could result in an increase in the intensity of sexual
selection on the sex that is more abundant. The OSR, which is the
average ratio of sexually active females to males in a population at
any given time, has been effective in estimating the opportunity for
sexual selection in several species (Vincent et al. 1994; Kvarnemo &
Ahnesjö 1996; Dearborn et al. 2001; Jones et al. 2001). However,
obtaining an accurate OSR for a population may not be feasible for
a particular species being studied and could also vary greatly
temporally (Forsgren et al. 2004), which could in turn have consequences for conclusions about the strength of sexual selection.
In situations where it is not practical to estimate the PI or OSR,
the potential reproductive rate can be used to predict the intensity
of sexual selection. The potential reproductive rate (PRR) is the rate
at which each sex in a population could reproduce if given unlimited mates (Clutton-Brock & Vincent 1991) and assumes no differential immigration or mortality between the sexes (Kokko &
Monaghan 2001). The PRR has proven to be a useful tool in predicting the direction and strength of sexual selection (fish:
0003-3472/$38.00 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd.
doi:10.1016/j.anbehav.2009.05.036
Author's personal copy
748
S.K. Scobell et al. / Animal Behaviour 78 (2009) 747–753
Kvarnemo 1994; Kvarnemo & Ahnesjö 1996; Masonjones & Lewis
2000; Ahnesjö et al. 2001; Wilson 2009; insects: Kvarnemo &
Simmons 1998; Wiklund et al. 1998; mammals: Preston et al. 2005).
It is usually measured under controlled laboratory conditions
where competition for mates is eliminated and environmental
variables are kept constant (i.e. temperature, food, shelter, lack of
predation, etc.). Thus, the PRR controls for many of the confounding
variables that can influence reproductive function and gives an
estimate of the baseline reproductive output for each sex under
a similar, constant environment.
PRR is typically measured in one of two ways: the number of
offspring (or potential offspring) produced per unit time (CluttonBrock & Vincent 1991; Ahnesjö et al. 2001; Wilson 2009) or the
‘time in/time out’ model (Clutton-Brock & Parker 1992). CluttonBrock & Vincent (1991) defined the PRR as the maximum number of
offspring that each sex produced per unit time. Later, Ahnesjö et al.
(2001) measured PRR by counting the number of ‘potential’
offspring produced per day over the course of the breeding season.
They argued that the female’s PRR should be based on the number
of eggs released during mating and that the male’s PRR should be
based on the number of eggs fertilized. For the ‘time in/time out’
model, Clutton-Brock & Parker (1992) defined ‘time out’ as the time
period in which an adult is not able to mate, and they defined ‘time
in’ as the time period when an animal is able to mate if given
a receptive partner. Both methods (‘time in/time out’ and number of
offspring/unit time) have proven useful and which method is used
often depends on the mating system of the species being studied.
The PRR has been used to study the mating system of several
species of syngnathids (pipefish and sea horses). The family Syngnathidae is unique among fishes in that male pregnancy is found
in all species and some, but not all species, are sex-role reversed
(Vincent et al. 1992). In species where the PRR of females exceeds
that of males, the mating system is typically sex-role reversed and
females compete for access to males. Nerophis ophidion and Syngnathus typhle females were estimated to be able to fill the pouch of
one male completely and still have a 41% surplus of eggs remaining
to give to another male (Berglund et al. 1989). Ahnesjö (1995) later
calculated that S. typhle had a female-biased of PRR 1.8:1 (females
to males) in warm water and 2.3:1 in cold water. Both N. ophidion
and S. typhle are pipefish that have sex-role reversed mating
systems in which females compete for access to males (Berglund &
Rosenqvist 2003). In species where the PRR of males and females
approaches unity, the mating system is typically monogamous with
conventional sex roles (but see Sogabe & Yanagisawa 2007). Vincent et al. (1994) were the first to show that despite their pregnancy, male Sri Lankan sea horses, Hippocampus fuscus, had
a higher PRR and displayed more competitive and courtship
behaviour than females. Similarly, Masonjones & Lewis (2000)
found that dwarf sea horses, Hippocampus zosterae, have a PRR of
1:1.2 and also show a conventional, monogamous mating system.
However, the PRR does not always predict a species’ mating
behaviour. Sogabe & Yanagisawa (2007) studied the PRR of the
monogamous messmate pipefish, Corythoichthys haematopterus,
and found that there was no difference in PRR between the sexes,
but females were more active in courtship than males. The
observed behavioural differences were associated with a femalebiased OSR for this population. Further studies are needed in this
family to determine how PRR is related to the mating system.
We conducted a study of the PRR of the sex-role reversed gulf
pipefish, Syngnathus scovelli, over a 2-month period. The gulf pipefish (Evermann & Kendall 1896) is a common syngnathid found
throughout the Gulf of Mexico and breeds year round in Florida,
U.S.A. (Reid 1954; Joseph 1957; Brown 1972). Gulf pipefish males,
like many members of the family Syngnathidae, have a pouch into
which the female deposits her eggs (her only known contribution).
The male fertilizes the eggs and then broods them for about 14 days
(Brown 1972). Thus, male gulf pipefish contribute significantly more
time to parental care than do females. The gulf pipefish is also
sexually dimorphic and has a polyandrous mating system. Females
are typically larger than males and have a sexually dimorphic
ornament, a silver bar on each bony ring of the trunk (Reid 1954).
Jones et al. (2001) found that mated females in the field had a larger
mean snout–vent length, body depth and body mass and were more
ornamented than unmated females. They also found that the standardized variance in mating success as measured by microsatellite
markers was at least seven times greater in females than in males.
Despite morphological and genetic evidence supporting
a female-biased PRR, it is not known whether male and female gulf
pipefish do indeed differ in their respective potential reproductive
rates. If the assumptions underlying sexual selection theory are
correct, they should be supported both in species with conventional mating systems and in those that are sex-role reversed
(Clutton-Brock 2009). The present study was designed to test the
hypothesis that the sex differences in body size, ornamentation and
parental roles of this species are accompanied by a sex difference in
PRR. The 2-month study period allowed us to measure PRR over
multiple breeding bouts. We predicted that females would have
a higher PRR than males and that the rates in both sexes would
change over successive breeding bouts in relationship to previous
reproductive output.
METHODS
Animals
We collected sexually mature S. scovelli adults (N ¼ 116) from
sea grass beds in Sarasota Bay, Florida on 6 June 2004. Sexually
mature males have a developed ventral brood pouch and females
have complete silver bars on the trunk (Brown 1972; Jones & Avise
1997). Pipefish were collected with push nets (1 m2, with 2 mm2
mesh) at a depth of roughly 1 m. Fish were shipped overnight to the
University of Oklahoma where they were housed in temporary,
nonreproductive groups in aquaria that did not share a common
water system. Most fish were allowed to acclimate to laboratory
conditions for 10 days before the study began. Five focal females
spent an additional 10–18 days acclimating while we waited for
a sufficient number of field-pregnant males to return to breeding
condition to pair with them.
Animal Housing
Fish were housed in 75-litre aquaria divided with opaque
Plexiglas either in half or into a small (18.75-litre) and a large
(56.25-litre) compartment (depending on treatment, see below).
Aquaria contained both biological sponge filters and undergravel
filters and were supplemented with saltwater bacteria once per
week. Aquaria contained crushed coral and sand substrate, plastic
plants that mimic natural sea grass (Thallassia sp.) and 15 cm long
by 3 cm wide sections of grey PVC tubing cut in half for cover. Fish
were fed newly hatched and adult Artemia enriched with Selco
(Aquatic Ecosystems, Apopca, FL, U.S.A.) twice daily. The feeding
regime was sufficient to maintain a body mass to total length (TL)
ratio that was similar to that found in the field (S.K.S., unpublished
data). Fish were kept at 25 C and on a 13:11 h light:dark cycle to
mimic natural conditions in Sarasota Bay during the summer.
Study Design
We conducted a 61-day study where we provided focal male
and female gulf pipefish with a surplus of mates and counted the
Author's personal copy
S.K. Scobell et al. / Animal Behaviour 78 (2009) 747–753
number of potential offspring each sex could produce. At the start
of the study, 12 male and 12 female focal subjects were chosen to
represent the range of body sizes for adult breeding individuals in
the population (TL ranges: males 119–145 mm, N ¼ 68; females
112–169 mm, N ¼ 48). Focal animals’ TL (134.7 2.6 mm) did not
differ from that of the population sample (132.4 1.2 mm; Mann–
Whitney U test: U ¼ 1067.0, N1 ¼ 24, N2 ¼ 92, P ¼ 0.58). Focal
animals were provided with similarly sized (within approximately
5 mm TL) opposite-sex partners because size-assortative mating
has been documented in several species of syngnathids (Vincent &
Sadler 1995; Jones et al. 2003; Silva et al. 2008).
The number of partner males or females that each sex could
mate with in 1 day was determined in a pilot study in the autumn of
2003. Females transferred eggs to the pouches of a maximum of
two males per day; males only accepted eggs from one female on
the first day of the pregnancy. In the current study, to ensure that
focal animals had a surplus of mates, each focal female was housed
with three breeding males and each focal male was housed with
two breeding females. Focal males were housed in larger
compartments (56.8 litres) than focal females (37.5 litres) to reduce
animal density and thus reduce female–female competition
(between a focal male’s two partner females; S.K.S., personal
observation) as this female competition was not the focus of the
present study and could have interfered with assessment of PRR.
On the four occasions when a focal male failed to get pregnant, one
of his partner females was removed (the suspected subordinate)
and replaced with a new partner female. This usually resulted in
a new pregnancy the next day (N ¼ 3). Male–male aggression has
not been observed in this species, so we were able to house the
focal females and their partner males in smaller compartments
than those used for the focal males and their partner females. As
the study progressed and the number of available breeding partner
males became limited (due to many males being pregnant), females
were housed with at least two breeding males, the maximum
number of males that a female could mate with in one day.
All focal and partner males were checked daily between 1600
and 2000 hours for new pregnancies. Pregnancies were easily
observed as the orange eggs are clearly visible though the skin of
the male’s pouch. The day that the pregnancy was detected was
designated day 0. Pregnant focal males remained in their home
tank with both females. Partner males, after becoming pregnant,
were moved to an adjacent 37.5-litre compartment, and a new
partner male in breeding condition was placed in with that focal
female. For all pregnant males (focal and partner), at mid-pregnancy we counted the eggs transferred by the female, and then,
following birth, we counted the total number of offspring. To obtain
the mid-pregnancy count, we counted the number of fertilized eggs
and any unfertilized eggs on or around day 7 of the pregnancy; at
this time, embryos are pigmented and fertilized eggs can be
distinguished from unfertilized eggs. The males were lightly
anaesthetized using a 0.05% solution of 2-phenoxyethanol (2-PE)
and the contents of the pouch were counted through the transparent skin folds. On day 12 of pregnancy, pregnant males were
placed in breeding nets containing a plastic plant and a PVC shelter
in their home aquaria. Breeding nets and the male’s pouch were
checked daily to determine whether the male had given birth. On
the day of birth, the pouch contents (newborn, underdeveloped
(dead) offspring and unfertilized eggs) were counted, newborn
were euthanized with an overdose of 2-PE, and all pouch contents
were then stored in 10% buffered formalin. After the 61-day study
period, we recorded TL, snout–vent length (SVL), body depth just
anterior to the dorsal fin, and total body mass for the focal male and
females. We euthanized focal males and females with an overdose
of 2-PE and dissected the gonads, which we preserved in 10%
formalin and later weighed to the nearest 0.1 mg.
749
Two focal females did not mate during the study, probably
because of reproductive suppression (regression of the gonads in
group-housed females; e.g. Rosenqvist 1990) during acclimation,
and we removed them from analyses. This left 10 females in all
analyses with the exception of analyses involving gonad mass. For
these analyses, sample sizes were reduced (males: N ¼ 10,
females: N ¼ 7) because of missing gonad data for several
animals.
The methods used in this study comply with the current laws of
the United States of America, in which the studies were performed,
and were approved by the University of Oklahoma Institutional
Animal Care and Use Committee (R02-013).
Analyses
We calculated the PRR by measuring the potential offspring
produced per unit time. The gulf pipefish has a courtship period
that can be extremely short. Under laboratory conditions, the time
from introduction of a partner to mating can be less than 1 min.
Thus, individual mating events can be easily missed without 24 h
observation, which was not logistically possible in this study. For
each subject, we calculated total fecundity over the study period.
To obtain the best measure of the focal animal’s fecundity for
each pregnancy, we compared the mid-pregnancy counts to the
pouch contents following birth, and then used whichever count
was higher in our analyses. This approach minimized the error of
the mid-pregnancy counts where some eggs, or occasionally an
entire row of eggs, could be occluded from the scorer’s view. We
calculated the PRR as each animal’s fecundity/day. However, as
PRR is a measure of an individual’s potential offspring, we
calculated the PRR for females as the number of eggs transferred/
day (for mid-pregnancy count) or all pouch contents (for count at
birth), and for males, we calculated the PRR as the number of
fertilized eggs/day (for mid-pregnancy count) or newborn þ
underdeveloped young (for count at birth), whichever was greater.
We used Student’s t test to determine whether there was a sex
difference in mean PRR between males and females for the entire
study period.
For the repeated measures analysis, we divided the PRR data
over the 61-day study into four breeding bouts. A breeding bout
was defined as the time that it took for males to complete a pregnancy (from mating to birth and thus return to breeding condition,
N ¼ 12, mean SE ¼ 15.0 1 days). The PRR for each of the four
breeding bouts was compared for males and females using
repeated measures mixed model analysis (see Brown & Prescott
1999). Because most populations of S. scovelli show sexual size
dimorphism (Reid 1954; Brown 1972; Jones et al. 2001), we used
a principal components analysis (using PC-ORD, MJM Software
Design, Gleneden Beach, OR, U.S.A.) to obtain a composite score for
each focal animal’s body mass, depth, TL and SVL. Axis 1 explained
91.6% of the total variance in focal animals’ body size. We fitted the
body size PCA axis 1 scores and subjects (nested within sex) as
random, so the effects of sex, breeding bout, PCA, and the interaction between sex*bout and sex*PCA on PRR could be assessed
across focal animals. We used least squares means with post hoc
Tukey–Kramer tests for sex and breeding bout.
To assess the relationship between morphological measurements and PRR, we used linear regression analysis to determine
whether TL, somatic mass (total mass gonad mass) or gonad
mass affected the PRR for each sex. We used correlation analysis to
determine the relationship between TL, somatic mass and gonad
mass for each sex.
The mixed model analysis was conducted using SAS 8.01 (SAS
Institute, Inc., Cary, NC, U.S.A.). All other analyses were conducted
using SPSS 15.0 (SPSS, Inc., Chicago, IL, U.S.A.).
Author's personal copy
750
S.K. Scobell et al. / Animal Behaviour 78 (2009) 747–753
RESULTS
Overall PRR
Females mated on average 2.8 times more often than males over
the course of our study (mean SE; females: 11.3 1.2, males:
4.1 0.2; Mann–Whitney U test: U ¼ 0.0, N1 ¼ 10, N2 ¼ 12,
P < 0.001; Fig. 1a). Accordingly, females had a shorter mean interbrood interval (number of days between matings ¼ 6.3 0.8 days)
than males (16.1 0.6 days, Student’s t test: t20 ¼ 10.09, P < 0.001;
Fig. 1b). The males’ longer interbrood interval was due mainly to the
length of pregnancy, rather than limited by female behaviour,
12
10
PRR and Body Size
(a)
Number of pregnancies
PRR over Four Breeding Bouts
The mean PRR of males and females did not remain constant
over the four breeding bouts (Fig. 2). A repeated measures mixed
model analysis showed that the interactions of sex*bout and
sex*PCA were not significant (sex*bout: F3,57 ¼ 0.52, P ¼ 0.67,
sex*PCA: F2,17 ¼ 2.18, P ¼ 0.14); these terms were removed from the
model and it was rerun. The main effects of sex and breeding bout
both affected PRR (sex: F1,20 ¼ 21.79, P ¼ 0.0001; bout: F3,63 ¼ 3.10,
P ¼ 0.03). Post hoc analyses showed that the PRR of breeding bout 1
was significantly greater than the PRR for breeding bout 2 (least
squares means with Tukey–Kramer adjustment: t63 ¼ 2.92,
P ¼ 0.005). The decrease in PRR from the first to second breeding
bout tended to be greater for females (4.0 1.4) than males
(1.9 0.7) (Student’s t test: t20 ¼ 4.28, P ¼ 0.052). The PRR for
breeding bout 2 was also significantly lower than the PRR for
breeding bout 3 (t63 ¼ 2.21, P ¼ 0.03). However, the change in PRR
between the second and third breeding bouts did not differ
significantly between females (3.2 1.0) and males (1.3 0.6)
(Student’s t test: t20 ¼ 0.70, P ¼ 0.41).
For both male and female focal animals, we noticed a common
pattern of fecundity over the four breeding bouts. Most individuals
had high fecundity during the first and third breeding bouts, but
much lower fecundity in the second breeding bout (see Fig. 3). Not
all individuals showed this pattern, but all showed at least one
‘boom’ and then ‘bust’ in fecundity over consecutive breeding bouts
during the study.
14
*
12
10
8
6
4
2
0
18
Interbrood interval (days)
because focal males were pregnant within 2.0 0.7 days of giving
birth (median ¼ 0.0 days; most males got pregnant the day after
giving birth). Females had a mean PRR 1.95 times greater than that
of males over the entire study period of 61 days. Focal females had
a mean PRR of 6.5 1.1 offspring/day, whereas focal males had
a mean PRR of 3.4 0.3 offspring/day (Student’s t test: t20 ¼ 2.99,
P ¼ 0.01; Fig. 1c).
16
(b)
*
14
8
Although the ranges of male and female body size measurements overlapped for the focal animals in this study, females in our
sample, as in other studied populations (Reid 1954; Joseph 1957;
6
4
2
12
0
PRR (offspring/day)
(c)
Mean PRR (offspring/day)
8
*
6
4
2
P = 0.005
10
8
6
4
2
1.90:1
0
0
Females
Males
Figure 1. Mean SE (a) number of pregnancies, (b) interbrood interval and (c)
potential reproductive rate (PRR) of female and male gulf pipefish over the 61-day
study. N ¼ 10 females, N ¼ 12 males. *P < 0.01.
P = 0.03
1
1.77:1
2.05:1
2
3
Breeding bout
2.03:1
4
5
Figure 2. Potential reproductive rates (PRR) for female (C) and male (B) gulf pipefish
over four breeding bouts (mean SE). The ratio of female to male PRR is shown for
each breeding bout. Brackets and associated P values denote differences in the PRR
between the indicated time periods.
Author's personal copy
140
16
120
14
PRR (offspring/day)
Fecundity (offspring/pregnancy)
S.K. Scobell et al. / Animal Behaviour 78 (2009) 747–753
100
80
60
40
12
10
8
6
4
2
20
0
751
21 Jun 2004
5 Jul 2004
0
50
19 Jul 2004 2 Aug 2004
Figure 3. Representative fecundity data from an individual female and an individual
male gulf pipefish. Each data point represents a single pregnancy over the course of the
study. Female 19 (C) and male 11 (B) best exemplified the ‘boom’ and ‘bust’ mating
patterns observed in the focal animals as well as the sex differences in mating
frequency.
Brown 1972), were generally larger than males (Table 1). However,
neither female TL (linear regression: R2 ¼ 0.01, P ¼ 0.81) nor
somatic mass (R2 < 0.01, P ¼ 0.91) was related to the PRR. Ovarian
mass also did not correlate with female TL (Spearman rank correlation: rS ¼ 0.07, N ¼ 7, P ¼ 0.88) or somatic mass (rS ¼ 0.11, N ¼ 7,
P ¼ 0.82), but females with a greater ovarian mass also had a higher
PRR (R2 ¼ 0.77, P ¼ 0.01) (Fig. 4). For males, testes mass increased
with both TL (Spearman rank correlation: rS ¼ 0.94, N ¼ 10,
P < 0.01) and somatic mass (rS ¼ 0.95, N ¼ 10, P < 0.01). However,
larger males and males with larger testes did not have a greater PRR
(TL: R2 < 0.01, P ¼ 0.82; somatic mass: R2 < 0.01, P ¼ 0.99; testes
mass: R2 < 0.01, P ¼ 0.96).
DISCUSSION
Our study documented that the PRRs of male and female gulf
pipefish differ, with females having an overall PRR almost twice
that of males. As in other sex-role reversed species of syngnathids
(Berglund et al. 1989; Kvarnemo & Ahnesjö 1996), female gulf
pipefish are able to mate more often and with a shorter interbrood
interval than are males. The gulf pipefish mating system is
consistent with sexual selection theory in that the sex that has the
higher PRR, females, also has the higher variance in mating success
(Jones et al. 2001) and is the more ornamented sex (Reid 1954;
Brown 1972; Jones et al. 2001; Clutton-Brock 2009). Assuming a 1:1
adult sex ratio and no differential mortality, we predict that gulf
pipefish have a female-biased OSR in the field (Jones & Avise 1997;
Ahnesjö et al. 2001; Jones et al. 2001) and, accordingly, that females
Table 1
Comparison of morphological measurements of focal male and female gulf pipefish
following the study period of 61 days (mean SE)
Females
10
59.82.0
140.24.8
7.20.3
2.30.2
116.930.9
Males
12
49.41.0
127.62.5
5.50.2
1.80.1
5.40.5
*P < 0.05; **P < 0.01.
y
Sample size for gonad mass: females N ¼ 7, males N ¼ 10.
150
200
250
300
Ovary mass (mg)
Date of mating
Sample size (N)
Snout–vent length (mm)
Total length (mm)
Depth of trunk (mm)
Body mass (g)
Gonad mass (mg)y
100
t
P
5.01
2.44
4.45
1.98
4.37
<0.01**
0.02*
<0.01**
0.06
0.01*
Figure 4. Potential reproductive rate (PRR) versus ovarian mass for focal gulf pipefish
females.
court males and compete with each other for access to mates (Jones
& Avise 1997; Clutton-Brock 2009), as we have observed in the
laboratory.
Although it has been suggested previously (Ahnesjö et al. 2001),
the present study is, to our knowledge, one of the few studies to
measure PRR over several breeding bouts in a vertebrate. The PRR of
gulf pipefish males, and particularly females, varied over the four
breeding bouts. The mean PRR was highest for both sexes during
the first breeding bout. Then, both sexes showed a decrease in
mean PRR during the second breeding bout. This general pattern of
a boom followed by a bust in reproductive investment was seen for
most of our focal animals. The change in male and female PRR over
the study is consistent with life-history theory predictions that
individuals that mate multiply make trade-offs between current
and future reproductive investment given a limited lifetime energy
budget and potential availability of mates (Stearns 1992; Zera &
Harshman 2001; Roff 2002; Heubel et al. 2008). However, it is also
possible that this pattern of reproduction was a result of the laboratory environment. Future studies of this species in the field and in
the laboratory will help clarify the factors that contribute to the
variation in PRR that we observed across multiple bouts of mating.
Saltwater populations of gulf pipefish generally show sexual size
dimorphism (Reid 1954; Joseph 1957; Brown 1972). Our sample
from the Sarasota Bay population also showed sexual size dimorphism, with females, on average, larger than males. Larger female
body size is common in fish and is often related to a fecundity
advantage (Andersson 1994). When each sex was examined individually, neither male or female total length nor somatic mass
affected the PRR. However, males and females did differ with
respect to the relationship of gonad size to PRR. Females that had
larger ovaries had a higher PRR, whereas males with larger testes
showed no concomitantly greater PRR. A female with larger ovaries
probably has a greater fitness advantage over a female with smaller
ovaries. Females with larger ovaries can transfer more eggs, and
because males in this species typically only accept eggs from one
female (Jones & Avise 1997; Jones et al. 2001), they would be predicted to choose the female that can transfer the greater number of
eggs (Rosenqvist 1990). As a result of male pregnancy, males with
bigger testes probably do not gain a significantly greater fitness
advantage over males with smaller testes. Males in this species are
thought to fertilize the eggs once they are in the male’s pouch,
similar to fertilization in S. schlegeli (Watanabe et al. 2000). There
has been no record of multiple paternity in S. scovelli (Jones & Avise
1997; Jones et al. 2001). Thus, there appears to be no sperm
Author's personal copy
752
S.K. Scobell et al. / Animal Behaviour 78 (2009) 747–753
competition within the brood pouch and sperm probably do not
have to fertilize eggs in the external environment. This speculation
is also supported by the very small size of testes in this species. Our
results are consistent with sexual selection acting more strongly on
female ovary size, with relaxed selection on male testis size
(Clutton-Brock 2009).
The difference in PRR between the sexes in gulf pipefish is
presumably influenced by the differing constraints of their reproductive physiology. Males are constrained by the time that it takes
to brood offspring, whereas females are only constrained by the
time that it takes to produce another batch of mature eggs. In our
study, a male’s ‘time out’ was consistently 14–15 days for each
pregnancy. Females, in contrast, were able to remate rapidly. For
example, one female mated with eight males within a 2-week
period, delivering full clutches of eggs (enough to fill the male’s
brood pouch) to five of them. To be able to mate multiply in rather
rapid succession, females must have a reproductive system that
facilitates successive rounds of ovulation in a short time period. The
ovary of the gulf pipefish is unusual because it is structured in
a conveyor belt fashion (Begovac & Wallace 1987). As one set of
eggs is ovulated into the lumen, the next set moves into its place
and completes the final stage of maturation (Begovac & Wallace
1988). The anatomy suggests that female gulf pipefish may have
a very short period of time between the ovulation of each set of
oocytes. However, we are unaware of any published studies on the
length of the female reproductive cycle or the hormonal mediation
of maturation of oocytes and ovulation in this species.
Recent research does suggest a close link between ovarian
morphology and PRR within the family Syngnathidae. Sea horses
and the monogamous pipefish C. haematopterus have ovaries that
differ in morphology from those of the gulf pipefish. Recently,
Sogabe et al. (2008) found that C. haematopterus females have
ovaries with two germinal ridges, similar to ovaries in sea horses
(Selman et al. 1991), but not to ovaries of other pipefish in the
genus Syngnathus. These authors suggested that this structural
difference might relate to the functional difference in reproduction between polyandrous/polygynandrous Syngathus pipefishes
with female-biased PRRs versus the monogamous sea horses and
C. haematopterus pipefish with PRRs that approach unity. The
group-synchronous ovaries of the monogamous sea horses and
C. haematopterus result in a longer ‘time out’ for these females
(Hippocampus spp.: 9–45 days, Foster & Vincent 2004; C. haematopterus: 10–19 days, Sogabe et al. 2007) than for Syngnathus
females (48 h, Berglund et al. 1988; present study) that have an
asynchronous-type ovary (Wallace & Selman 1981; Begovac &
Wallace 1988). It is possible that these two types of ovarian
morphologies dictate the speed at which females of these species
can produce mature eggs, which would directly affect the PRR.
Further studies are needed to determine how the reproductive
physiology of the ovarian cycle relates to these two distinct
ovarian morphologies and whether ovarian morphology and/or
physiology produce functional constraints that affect the PRR in
these species.
Acknowledgments
We thank P.L. Schwagmeyer and Edith Marsh-Matthews for
their help with the analysis and comments on the manuscript. We
also thank Ingo Schlupp, Celeste Wirsig and Debra Bemben for help
with many aspects of this research, and Adam Jones, Anders
Berglund and an anonymous referee for helpful comments on an
earlier version of the manuscript. Navin Chowhudry, Joe Kelly,
Claire Lukeman, Amber Mielke and Lori Deunk Garman provided
much assistance with animal husbandry. We also thank Leila
Wright for assistance in the field. We are especially grateful to John
Stanley and Dave Jenkins at the Mote Marine Laboratory for assistance with animal collection and aquaculture techniques. Funding
to S.K.S. from the University of Oklahoma’s Graduate Student
Senate, the Department of Zoology’s Adams Memorial Scholarship,
Sigma Xi and the PADI Foundation supported this study. This study
also benefited from the funds awarded to R.K. from the University
of Oklahoma Research Council.
References
Ahnesjö, I. 1995. Temperature affects male and female potential reproductive rates
differently in the sex-role reversed pipefish, Syngnathus typhle. Behavioral
Ecology, 6, 229–233.
Ahnesjö, I., Kvarnemo, C. & Merilaita, S. 2001. Using potential reproductive rates
to predict mating competition among individuals qualified to mate. Behavioral
Ecology, 12, 397–401.
Andersson, M. 1994. Sexual Selection. Princeton, New Jersey: Princeton University
Press.
Begovac, P. C. & Wallace, R. A. 1987. Ovary of the pipefish, Syngnathus scovelli.
Journal of Morphology, 193, 117–133.
Begovac, P. C. & Wallace, R. A. 1988. Stages of oocyte development in the pipefish,
Syngnathus scovelli. Journal of Morphology, 197, 353–369.
Berglund, A. & Rosenqvist, G. 2003. Sex role reversal in pipefish. Advances in the
Study of Behavior, 32, 131–167.
Berglund, A., Rosenqvist, G. & Svensson, I. 1988. Multiple matings and paternal
brood care in the pipefish Syngnathus typhle. Oikos, 51, 184–188.
Berglund, A., Rosenqvist, G. & Svensson, I. 1989. Reproductive success of females
limited by males in two pipefish species. American Naturalist, 133, 506–516.
Brown, J. D. 1972. A comparative life history study of four species of pipefishes
(family Syngnathidae) in Florida. Ph.D. thesis, Univeristy of Florida, Gainesville.
Brown, H. & Prescott, R. 1999. Applied Mixed Models in Medicine. West Sussex:
J. Wiley.
Clutton-Brock, T. H. 1991. The Evolution of Parental Care. Princeton, New Jersey:
Princeton University Press.
Clutton-Brock, T. H. 2009. Sexual selection in females. Animal Behaviour, 77, 3–11.
Clutton-Brock, T. H. & Parker, G. A. 1992. Potential reproductive rates and the
operation of sexual selection. Quarterly Review of Biology, 67, 437–456.
Clutton-Brock, T. H. & Vincent, A. C. J. 1991. Sexual selection and the potential
reproductive rates of males and females. Nature, 351, 58–60.
Dearborn, D. C., Anders, A. D. & Parker, P. G. 2001. Sexual dimorphism, extrapair
fertilizations, and operational sex ratio in great frigatebirds (Fregata minor).
Behavioral Ecology, 12, 746–752.
Emlen, S. T. & Oring, L. W. 1977. Ecology, sexual selection, and the evolution of
mating systems. Science, 197, 215–223.
Evermann, B. W. & Kendall, W. C. 1896. Description of a new species of pipefish
(Siphostoma scovelli) from Corpus Christi, Texas. Proceedings of the U.S. National
Museum, 18, 113–115.
Forsgren, E., Amundsen, T., Borg, A. A. & Bjelvenmark, J. 2004. Unusually dynamic
sex roles in a fish. Nature, 3429, 551–554.
Foster, S. J. & Vincent, A. C. J. 2004. Life history and ecology of seahorses: implications for conservation and management. Journal of Fish Biology, 65, 1–61.
Heubel, K. U., Lindström, K. & Kokko, H. 2008. Females increase current reproductive effort when future access to males is uncertain. Biology Letters, 4,
224–227.
Jones, A. G. & Avise, J. C. 1997. Microsatellite analysis of maternity and the mating
system in the gulf pipefish Syngnathus scovelli, a species with male pregnancy
and sex-role reversal. Molecular Ecology, 6, 203–213.
Jones, A. G., Walker, D. & Avise, J. C. 2001. Genetic evidence for extreme polyandry
and extraordinary sex-role reversal in a pipefish. Proceedings of the Royal Society
B, 268, 2531–2535.
Jones, A. G., Moore, G. I., Kvarnemo, C., Walker, D. & Avise, J. C. 2003. Sympatric
speciation as a consequence of male pregnancy in seahorses. Proceedings of the
National Academy of Sciences, U.S.A., 100, 6598–6603.
Joseph, E. B. 1957. A study of the systematics and life history of the gulf pipefish
Syngnathus scovelli (Evermann and Kendall). Ph.D. thesis, Florida State University, Tallahassee.
Kokko, H. & Monaghan, P. 2001. Predicting the direction of sexual selection.
Ecology Letters, 4, 159–165.
Kvarnemo, C. 1994. Temperature differentially affects male and female reproductive rates in the sand goby: consequences for operational sex ratio. Proceedings
of the Royal Society B, 256, 151–156.
Kvarnemo, C. & Ahnesjö, I. 1996. The dynamics of operational sex ratios and
competition for mates. Trends in Ecology & Evolution, 11, 404–408.
Kvarnemo, C. & Simmons, L. W. 1998. Male potential reproductive rate influences
mate choice in a bushcricket. Animal Behaviour, 55, 1499–1506.
Masonjones, H. D. & Lewis, S. M. 2000. Differences in potential reproductive rates
of male and female seahorses related to courtship roles. Animal Behaviour, 59,
11–20.
Preston, B. T., Stevenson, I. R., Pemberton, J. M., Coltman, D. W. & Wilson, K.
2005. Male mate choice influences promiscuity in Soay sheep. Proceedings of the
Royal Society B, 272, 365–373.
Author's personal copy
S.K. Scobell et al. / Animal Behaviour 78 (2009) 747–753
Reid, G. K. 1954. An ecological study of the Gulf of Mexico fishes, in the vicinity of
Cedar Key, Florida. Bulletin of Marine Science of the Gulf and Caribbean, 4, 1–94.
Roff, D. A. 2002. Life History Evolution. Sunderland, Massachusetts: Sinauer.
Rosenqvist, G. 1990. Male mate choice and female–female competition for mates in
the pipefish Nerophis ophidian. Animal Behaviour, 39, 1110–1115.
Selman, K., Wallace, R. A. & Player, D. 1991. Ovary of the seahorse, Hippocampus
erectus. Journal of Morphology, 209, 285–304.
Silva, K., Vieira, M. N., Almada, V. C. & Monteiro, N. M. 2008. Can the limited
marsupium space be a limiting factor for Syngnathus abaster females? Insights
from a population with size-assortative mating. Journal of Animal Ecology, 77,
390–394.
Sogabe, A. & Yanagisawa, Y. 2007. Sex-role reversal of a monogamous pipefish
without higher potential reproductive rate in females. Proceedings of the Royal
Society B, 274, 2959–2963.
Sogabe, A., Matsumoto, K. & Yanagisawa, Y. 2007. Mate change reduces the
reproductive rate of males in a monogamous pipefish Corythoichthys haematopterus: the benefit of long-term pair bonding. Ethology, 113, 764–771.
Sogabe, A., Matsumoto, K., Ohashi, M., Watanabe, A., Takata, H.,
Murakami, Y., Omori, K. & Yanagisawa, Y. 2008. A monogamous pipefish
has the same type of ovary as observed in monogamous seahorses. Biology
Letters, 4, 362–365.
753
Stearns, S. C. 1992. The Evolution of Life Histories. Oxford: Oxford University Press.
Trivers, R. L. 1972. Parental investment and sexual selection. In: Sexual Selection and
the Descent of Man (Ed. by B. Campbell), pp. 136–179. Chicago: Aldine.
Vincent, A. C. J. & Sadler, L. M. 1995. Faithful pair bonds in wild seahorses,
Hippocampus whitei. Animal Behaviour, 50, 1557–1569.
Vincent, A. C. J., Ahnesjö, I., Berglund, A. & Rosenqvist, G. 1992. Pipefishes and
seahorses: are they all sex role reversed? Trends in Ecology & Evolution, 7, 237–241.
Vincent, A. C. J., Ahnesjö, I. & Berglund, A. 1994. Operational sex ratios and
behavioral sex differences in a pipefish population. Behavioral Ecology and
Sociobiology, 34, 435–442.
Wallace, R. A. & Selman, K. 1981. Cellular and dynamic aspects of oocyte growth in
teleosts. American Zoologist, 21, 325–343.
Watanabe, S., Hara, M. & Watanabe, Y. 2000. Male internal fertilization and
introsperm-like sperm of the seaweed pipefish (Syngnathus schlegeli). Zoological
Science, 17, 759–767.
Wiklund, C., Kaitala, A. & Wedell, N. 1998. Decoupling of reproductive rates and
parental expenditure in a polyandrous butterfly. Behavioral Ecology, 9, 20–25.
Wilson, A. B. 2009. Fecundity selection predicts Bergmann’s rule in syngnathid
fishes. Molecular Ecology, 18, 1263–1272.
Zera, A. J. & Harshman, L. G. 2001. Physiology of life history trade-offs in animals.
Annual Review of Ecology and Systematics, 32, 95–126.
`