Animal Behaviour 77 (2009) 465472

Contents lists available at ScienceDirect

Animal Behaviour
journal homepage: www.elsevier.com/locate/yanbe

Male monarch butteries, Danaus plexippus, adjust ejaculates

in response to intensity of sperm competition
M.J. Solensky a, *, K.S. Oberhauser b,1
a
b

Department of Biology, The College of Wooster

Department of Fisheries, Wildlife and Conservation Biology, University of Minnesota

a r t i c l e i n f o
Article history:
Received 29 July 2008
Initial acceptance 1 September 2008
Final acceptance 17 October 2008
Published online 11 December 2008
MS. number: A08-00497R
Keywords:
apyrene sperm
Danaus plexippus
eupyrene sperm
mate assessment
monarch buttery
sperm competition
sperm transfer
spermatophore
strategic mating effort

During mating, male Lepidoptera transfer spermatophores that consist of accessory gland material,
eupyrene (nucleated) sperm and apyrene sperm that is incapable of fertilizing eggs. Sexual selection
theory predicts that males should allocate these materials strategically based on the risk and intensity of
sperm competition. We studied the relationship between behavioural and physiological cues and
material allocation by male monarch butteries, Danaus plexippus. Males that had waited longer between
matings transferred larger spermatophores and more apyrene and eupyrene sperm. Eupyrene sperm
number was also correlated with female mating history, with males transferring more sperm to females
that had larger amounts of spermatophore material stored from previous mates, regardless of whether
this came from one or three mates. This result suggests that males use stored ejaculates to assess female
mating history and increase eupyrene sperm investment under increased sperm competition intensity.
Male monarchs appear to be capable of independently manipulating the different components of their
ejaculates. Ejaculate allocation patterns suggest that males benet by maximizing spermatophore size
and apyrene sperm number, possibly to delay future female remating. However, males allocate more
eupyrene sperm to females when sperm competition is more intense, which is consistent with predictions from recent sperm competition models.
2008 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved.

Studies of sexual selection traditionally focused on precopulatory processes, but in the past few decades researchers have
recognized that both female choice and male competition continue
after copulation (Parker 1970; Eberhard 1996). Sperm competition
occurs when the ejaculates of two or more males are simultaneously available to a female for fertilization. This competition
often approximates a rafe in which males delivering more sperm
are likely to fertilize more eggs (Parker 1990; Gage & Morrow
2003). However, the energetic demands of producing large quantities of sperm can limit male reproductive success (Dewsbury
1982; Wedell et al. 2002) and males are consequently predicted to
allocate ejaculates strategically based on male condition, risk and
intensity of sperm competition and relative reproductive value of
a mate (Parker 1990; Parker et al. 1996, 1997; Engqvist & Reinhold
2006, 2007; Ball & Parker 2007). Recent models distinguish
between sperm competition risk (i.e. the probability of sperm
competition occurring) and intensity (i.e. the number of competing

* Correspondence: M. J. Solensky, Department of Biology, The College of Wooster,

931 College Mall, Wooster, OH 44691, U.S.A.
E-mail address: [email protected] (M.J. Solensky).
1
K. S. Oberhauser is at the Department of Fisheries, Wildlife and Conservation
Biology, 1980 Folwell Ave., University of Minnesota, St Paul, MN 55108, U.S.A.

sperm or ejaculates). Theory predicts that males should typically

invest more sperm when the risk of sperm competition is high,
such as situations in which females have already mated or are likely
to remate (Parker et al. 1997; Engqvist & Reinhold 2006; but see Ball
& Parker 2007). Support for a positive relationship between the risk
of sperm competition and investment in sperm has been documented in many taxa (Gage 1991; Cook & Gage 1995; Birkhead &
Mller 1998; Wedell & Cook 1999a; Wedell et al. 2002; Evans et al.
2003; Zbinden et al. 2003; delBarco-Trillo & Ferkin 2004; GarcaGonzalez & Gomendio 2004; Pound & Gage 2004; Ramm & Stockley
2007; Thomas & Simmons 2007). Predictions of the effect of the
intensity of sperm competition on male investment are less
straightforward. Early models predicted that males should invest
more in sperm when the intensity of sperm competition is
moderate (e.g. a female has mated once), but not high (e.g. a female
mated multiple times; Parker et al. 1996). However, more recent
models developed for internally fertilizing animals predict more
complicated outcomes (Engqvist & Reinhold 2006, 2007). Male
response to the risk and intensity of sperm competition can vary
with the degree of polyandry (Yamane & Miyatake 2005), male
sperm reserves (Engqvist & Reinhold 2007), sperm precedence
patterns (Engqvist & Reinhold 2006; Engqvist 2007) and sperm
limitation in females (Ball & Parker 2007).

0003-3472/$38.00 2008 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.anbehav.2008.10.026

466

M.J. Solensky, K.S. Oberhauser / Animal Behaviour 77 (2009) 465472

Sperm precedence patterns vary considerably across taxa, from

complete rst-male precedence to complete last-male precedence,
with mixed paternity found in a number of species (reviewed in
Birkhead & Mller 1998). When the last male to mate has a strong
fertilization advantage, selection should favour males that prevent
or delay female remating, in some cases by chemical deposition
(Andersson et al. 2000, 2003; Grillet et al. 2006) or increased
ejaculate expenditure (Oberhauser 1989, 1992; Cook & Wedell
1999). Selection can also favour male competition against previous
mates, in some cases by increased male sperm investment in
matings with previously mated females (Engqvist & Reinhold
2006). Mixed paternity can selectively favour any combination of
these male strategies.
We studied patterns of ejaculate allocation in monarch butteries, Danaus plexippus. Male monarchs use internal fertilization and
obtain mates through coercion rather than courtship (Frey 1999;
Oberhauser & Frey 1999), which limits opportunities for precopulatory sexual selection. Regardless of the time of day at which
copulation begins, mating pairs typically remain coupled until after
dusk, and copulation ends only when the male releases the female
(Oberhauser & Frey 1999). During copulation, monarch males
transfer large spermatophores (up to 10% of body mass: Oberhauser
1988) that contain sperm and accessory gland material. Males
transfer accessory gland material throughout the day, but do not
initiate sperm transfer until after dusk (Svard & Wiklund 1988).
Copulation generally ends between 2200 and 0400 hours (M. Solensky, unpublished data). Shortly after copulation ends, sperm
move from the bursa copulatrix, where the spermatophore is
deposited, through the sperm duct and into the spermatheca,
where they are in position to fertilize eggs. The rest of the spermatophore remains in the bursa copulatrix for several days, where
it is partially digested by the female and causes a delay in remating
that is positively correlated with its size (Oberhauser 1989, 1992).
Because the last male to mate usually has a fertilization advantage
(Solensky 2003; Solensky & Oberhauser, in press), males benet
from delivering a large spermatophore by delaying female remating and consequently preserving last-male sperm advantage.
Spermatophore size depends primarily on male mating history;
spermatophore size in males increases with time since last
copulation (Oberhauser 1988).
Male Lepidoptera transfer both eupyrene sperm, capable of
fertilizing eggs, and apyrene sperm, which are smaller and lack
a nucleus. Males eclose with a nearly complete complement of
sperm (Seth et al. 2002), although sperm transfer from the testes to
a storage organ accessible during copulation occurs throughout the
males lifetime and appears to follow a daily rhythm (Bebas et al.
2001). Males appear to have some control over how many sperm
they transfer during mating: they may adjust sperm transfer based
on their own condition or their female mates condition (Wedell &
Cook 1999a) or their perceived risk of sperm competition (Cook &
Gage 1995; Wedell & Cook 1999a, b).
While there is abundant evidence that males can often adjust
their ejaculates in sophisticated ways (e.g. Pizzari et al. 2003;
Thomas & Simmons 2007), the mechanisms that males use to
assess sperm competition risk and intensity and to adjust ejaculates
often remain unclear (Wedell et al. 2002; Gage 2003). Males may
use female behaviour, male- or female-derived pheromones
(Siva-Jothy & Stutt 2003; Carazo et al. 2004), the presence of rival
males (Evans et al. 2003; Zbinden et al. 2003; Pound & Gage 2004;
Ramm & Stockley 2007), female mating history (Cook & Gage 1995;
Wedell 1998; Wedell & Cook 1999a) or characteristics of rival
ejaculates (Cook & Gage 1995; Wedell & Cook 1999a; Engqvist
2007) to assess the risk and intensity of sperm competition.
We investigated three categories of cues that might inuence
monarch sperm allocation. We measured behavioural cues (mating
attempt duration, copulation duration, timing of initiation and

termination of copulation) as potential indicators of female resistance and therefore mating history (Frey 1999), female condition
cues (size and mating history) and male condition cues (size and
mating history). To further investigate the effect of perceived
intensity of sperm competition on male ejaculate characteristics,
we manipulated female mating history such that females had
either small or large quantities of spermatophore material from
either one or three previous mates. If males use female bursa
copulatrix content as an indicator of female mating history and
subsequent intensity of sperm competition from previous mates,
then males should adjust eupyrene sperm transfer in response to
the quantity of spermatophore material stored, regardless of the
number of a females previous mates.
METHODS
Study Organisms
The adult monarchs used were the offspring of eggs and larvae
collected from the wild in Missouri, Arkansas and Ohio, U.S.A.
Laboratory colonies were derived from at least 20 founding
individuals. We reared larvae in screen cages (0.5  0.3  0.6 m;
50150 larvae per cage) on a mixture of potted Asclepias curassavica
and cut stems of A. syriaca. Larvae developed at room temperature
(roughly 24  C) under natural light conditions (roughly 18:6 h
light:dark cycle). On eclosion, we checked adults for infection by
Ophryocystis elektroscirrha (Altizer et al. 2000), labelled uninfected
adults on the hindwings using a permanent marker, and measured
forewing length or area. Adults were fed a 20% honeywater
solution to satiation every other day before being released into
cages. In the cages, they had continuous access to potted A. curassavica or cut stems of A. syriaca and a 20% honeywater solution.
Some females were stored in a cold chamber (12  C) for 510 days
before being used in the experiment. This treatment does not affect
their mating propensity or fecundity (K. Oberhauser, unpublished
data).
Observational Study: Mating History, Mating Behaviour
and Ejaculate Characteristics
We released an equal number of virgin males and virgin females
(510 days old, 30 of each sex) into each of six outdoor mesh cages
(1.8 m3) in St Paul, MN, and monitored mating activity for 10 days
starting on 10 July 2003. In 2006, we released an approximately
equal number of virgin males and virgin females (515 days old,
1020 of each sex) into each of seven mesh cages (0.6 m3) near
large windows in a warm laboratory (29  C) in Wooster, OH and
monitored mating activity for 19 days starting on 26 June.
In 2003, we observed the mating behaviour of most mating
attempts that led to the transfer of the ejaculates whose characteristics we measured. On 6 of the 10 days of the experiment, we
made detailed observations of the mating behaviour of the adults in
a focal cage: each cage was observed only once, and the order of
observation was randomly assigned. For each mating attempt
observed, we recorded the identity of the individuals involved, start
time, attempt duration, and whether the attempt resulted in
copulation. We recorded resistance behaviours (Frey 1999) used
by the female, including arc (female exes abdomen dorsally, wings
horizontal or folded), inverted (dorsal surface of female facing the
ground), sheath (females wings folded, usually the female lies
tilted to one side against substrate) and curl (females abdomen
curled forward under thorax, female often wraps her legs around
her curled abdomen).
To measure male ejaculate transfer, we dissected all mated
females in the focal cage (2003) or in all cages (2006). Measuring
the ejaculate transferred to an individual female was a destructive

M.J. Solensky, K.S. Oberhauser / Animal Behaviour 77 (2009) 465472

process, which ensured that no females were sampled twice. When

the ejaculates of individual males were collected from more than
one female, we used only the rst observation of each male in our
analyses; thus, all ejaculates were independent samples.
Experimental Study: Manipulated Female Mating History
and Ejaculate Characteristics
We generated three groups of females that differed in the
number and size of spermatophores received from mates prior to
mating with a focal male; females had received either one small (S)
spermatophore, one large (L) spermatophore, or three small (SSS)
spermatophores. Small spermatophores were delivered by males
that had mated 1 day prior, while large spermatophores were
delivered by males that had waited 4 days since last mating. This
difference in male mating history generates a three- to four-fold
difference in size between small and large spermatophores (Oberhauser 1988), which suggests that females in the L and SSS treatments should have stored similar total amounts of spermatophore
material prior to their focal matings. Males were randomly assigned
to deliver either a small or a large spermatophore, and on the day
after mating were either used to deliver a small (S) spermatophore
or removed from females for 4 days before being released into a cage
with experimental females to deliver a large (L) spermatophore.
Virgin females were randomly assigned to one of the three
mating history treatments, and then released into mesh cages
(0.6 m3) near large windows in a warm laboratory (29  C) in
Wooster, OH, that contained 1020 females from the same
treatment group. Females had at least 1 day without copulation in
between each mating, during which time they were removed to
a female-only holding cage. Upon receiving the appropriate
number and size of spermatophore based on the pre-assigned
treatment, females were held for 1 day without access to males and
then moved to a cage that housed males that had mated 1 day prior.
These were the focal males whose ejaculate characteristics we
measured in response to manipulated female mating history. This
cage housed all females from all three treatments that had previously received one or three large or small spermatophores. We kept
the sex ratio in this cage equal or male biased to maximize the
likelihood of all experimental females mating soon after they had
completed the preliminary matings, before the spermatophore
material would begin to break down (Oberhauser 1992).
Measuring Sperm Transfer
To measure the spermatophore size and number of sperm
transferred, we modied methods described by Oberhauser (1988)
and Cook & Wedell (1996). We monitored the cages every 15 min
from 2100 hours until the last pair had separated. When a pair
separated, we immediately collected the female and decapitated
and dissected it in insect saline within 10 min, before sperm had
left the spermatophore. We measured spermatophore mass by
dissecting out the spermatophore from the bursa copulatrix,
blotting it on tissue paper to uniform dryness and weighing it to the
nearest 0.1 mg on a Mettler semi-micro analytical balance. When
more than one spermatophore was present, we identied the
newest spermatophore by its light coloration and the presence of
sperm. In 2006, we also weighed the old spermatophore material
recovered from the females bursa copulatrix. We then dissected
the sperm sac from the newest spermatophore, transferred it to
a depression slide, ruptured it into a 10 ml drop of insect saline using
ne-point forceps, and counted the number of eupyrene sperm
bundles under 40 magnication. Among the Lepidoptera, each
eupyrene sperm bundle results from eight cell divisions (Virkki
1969), so we multiplied the number of bundles by 256 (28) to
determine the number of eupyrene sperm transferred.

467

We rinsed the sperm solution from the depression slide and

cover slip with deionized water into a tared 100 ml glass jar. We
diluted the sperm solution to approximately 60 ml with deionized
water and recorded the exact volume. After adding 10 ml of
toluidene blue protein stain, we gently agitated the solution and
transferred 10 ml to each of ve clean microscope slides. After
allowing the slides to dry, we dipped each in deionized water to
dissolve salt crystals remaining from the insect saline solution
without removing sperm (Cook & Wedell 1996). We scanned each
slide at 100 magnication using a compound light microscope to
count the number of apyrene sperm, then calculated the total
number of apyrene sperm by multiplying the average number of
sperm subsampled by its dilution factor.

Statistical Analyses
Three of the independent variables (male and female wing area,
and copulation duration) were normally distributed (Kolmogorov
Smirnov test: P > 0.5). Mating attempt duration was not normally
distributed, so we used the natural log of attempt duration in our
analyses, which did not differ statistically from a normal distribution (KolmogorovSmirnov test: P 0.489). Male time since last
mating (days) and female mating history (number of mates) were
not normally distributed (KolmogorovSmirnov test: P < 0.001),
and because both were measured as ordinal categories with a large
number of observations in the rst category, neither could be
transformed to approximate a normal distribution. Bursa copulatrix
content (the mass of all old spermatophores) was also not normally
distributed (KolmogorovSmirnov test: P 0.002) and could not be
transformed to approximate a normal distribution because of the
large number of observations of virgin females in the observational
study, for which this measure equalled zero. Two of the dependent
variables (spermatophore mass and the number of apyrene sperm)
were not normally distributed (KolmogorovSmirnov test:
P < 0.05), so we used the square root of spermatophore mass and
apyrene sperm number; these and eupyrene sperm number were
all normally distributed (KolmogorovSmirnov test: P > 0.15).
We used multiple linear regression to test for relationships
between ejaculate components and the four independent variables
that were measured in both years of this study (male and female
mating history and wing area) and separately for the behavioural
variables measured in 2003 only. Sperm competition models
predict that males should invest maximally in ejaculate transfer at
intermediate levels of sperm competition intensity (Parker et al.
1996; Engqvist & Reinhold 2007), which predicts a quadratic effect
of female mating history on ejaculate transfer, so we initially
included the quadratic term of female mating history in our
regression models. The quadratic term was not signicant in any of
the models (P > 0.05), so we report models with linear terms only.
Although male and female mating history did not conform to
a normal distribution, we included them in the regression models;
when either was identied as a signicant predictor, we assessed
the signicance of the pattern further using a nonparametric
Spearman rank correlation test. Within the female mating history
treatment groups, all three ejaculate components were normally
distributed (KolmogorovSmirnov test: P > 0.25). We used oneway ANOVA models to measure differences in each ejaculate
component between treatment groups. Because the three ANOVAs
measured different ejaculate characteristics (spermatophore mass,
eupyrene and apyrene sperm number), we applied a Bonferroni
adjustment and concluded statistical signicance when P < 0.017
(0.05/3) to maintain an experiment-wise error rate of P < 0.05
(Sokal & Rohlf 1995) for analyses that were repeated for these three
dependent variables. Unless otherwise noted, all estimates are
reported as mean  SE.

468

M.J. Solensky, K.S. Oberhauser / Animal Behaviour 77 (2009) 465472

RESULTS

Table 1
Regression models of ejaculate characteristics for square root of spermatophore
mass (N 131), square root of the number of apyrene sperm transferred (N 122)
and number of eupyrene sperm transferred (N 124)

Observational Study: Mating Behaviour

Mating attempts typically began with the male on top of the
female and both butteries dorsal side up (46 of 55 attempts for
which starting position was recorded). Of the four female resistance
behaviours recorded, the arc behaviour was the most common,
although it was only used in 16 of 51 mating attempts for which
behaviour was recorded. Only 4 of 51 females used the curl
behaviour, and all of these females had mated previously. In two
mating attempts, females used the inverted behaviour, and 4 of 51
females used the sheath behaviour. Because resistance behaviours
occurred so infrequently, they were not included as potential
predictors of spermatophore size or sperm transfer in the statistical
analyses. Mating attempts were longer when at least one resistance
behaviour was used (retransformed mean (95% condence interval,
CI): with resistance 84 (51,139) s; without resistance 25
(16, 37) s; t49 4.014, P < 0.001). Female mating history had only
a marginally signicant effect on whether a resistance behaviour was
used (logistic regression: N 51; model c21 2.671, P 0.102), but
previously mated females had longer mating attempts than virgins
(retransformed mean (95% CI): mated 55 (35, 84) s; virgin 29
(18, 46) s; t55 2.032, P 0.047). Females with more previous
matings had longer mating attempts (Spearman rank: rS 0.284,
N 57, P 0.032; Fig.1), although this pattern may be driven by four
females with three and four previous mates (Spearman rank
omitting potential outliers: rS 0.186, N 53, P 0.182).

Observational Study: Ejaculate Characteristics

In both years, mating pairs separated between 2140 and 0610
hours. The average spermatophore mass was 15.6  0.9 mg (range
1.240.8, N 156). On average, males transferred 81082  2683
(range 1280251136, N 146) eupyrene sperm and 1561693 
65 939 (range 290 4005 426 419, N 144) apyrene sperm.
Nonvirgin male time since last mating was the only variable
that affected spermatophore size (Spearman rank: rS 0.430,
N 132, P < 0.001; Table 1, Fig. 2a) and apyrene sperm transfer
(rS 0.398, N 123, P < 0.001; Table 1, Fig. 2c). Both male and
female mating history affected eupyrene sperm transfer (Table 1).
Males that had waited longer since the last mating transferred
larger spermatophores, and more eupyrene and apyrene sperm
(Fig. 2). Males transferred more eupyrene sperm to females that
had previously mated more frequently (Fig. 3a). Male and female

Log (attempt duration) (s)

1
2
3
Number of female's previous mates

Figure 1. Relation between female mating history and mating attempt duration.
Potential outliers shown as open circles.

Coefcient (SE)

Spermatophore mass
Constant
Male mating history
Female mating history
Male wing area
Female wing area

0.071 (0.037)
0.008 (0.001)
(0.0000.002)
(0.0000.000)
(0.0000.000)

<0.001
0.924
0.181
0.765

Apyrene sperm number

Constant
Male mating history
Female mating history
Male wing area
Female wing area

925.8 (434.8)
72.6 (13.5)
44.4 (24.8)
0.31 (0.43)
0.24 (0.43)

<0.001
0.076
0.480
0.575

Eupyrene sperm number

Constant
Male mating history
Female mating history
Male wing area
Female wing area

106675 (45664)
7983 (1413)
7085 (2608)
18 (45)
41 (45)

<0.001
0.008
0.699
0.364

Male mating history days elapsed since a nonvirgin males last mating; female
mating history number of previous copulations.

wing area, the time at which a pair coupled or separated, mating

attempt duration and copulation duration had no signicant
effects on any of the three ejaculate characteristics measured, and
female mating history had no effect on spermatophore mass or
apyrene sperm transfer (multiple linear regression: all predictors
P > 0.05).
The same patterns emerged from regression analyses using the
total mass of old spermatophores from previous copulations as
a measure of female mating history rather than the number of
female matings: spermatophore mass and apyrene sperm number
were correlated only with male mating history (multiple linear
regression: P < 0.001), while eupyrene sperm transfer was correlated with both male mating history (P < 0.001) and the total mass
of old spermatophores in the female (P 0.009; Fig. 3b).
Experimental Study: Female Mating History Manipulations
As expected, females in the three treatment groups differed in
the amount of old spermatophore material stored from previous
copulations (ANOVA: F2,39 10.295, P < 0.001; Fig. 4a). Females
that had previously received one small (S) spermatophore had less
old spermatophore material stored than females that had previously received one large (L) spermatophore (Tukey post hoc test:
P < 0.001) or three small (SSS) spermatophores (Tukey post hoc:
P 0.051). Females in the L group had marginally more old
spermatophore material stored than females in the SSS group
(Tukey post hoc: P 0.079).
Males transferred signicantly more eupyrene sperm to females
in the L and SSS groups than to females in the S group (ANOVA:
F2,37 5.28, P 0.010; Fig. 4b). Males transferred marginally more
apyrene sperm (ANOVA: F2,41 3.19, P 0.052; Fig. 4c) and smaller
spermatophores (ANOVA: F2,37 3.76, P 0.033; Fig. 4d) to
females in the SSS treatment than to females in the S treatment
(using the Bonferroni corrected P < 0.017 to conclude statistical
signicance), while females in the L treatment received intermediate amounts of apyrene sperm and spermatophore sizes.
DISCUSSION
Our results suggest that male monarch butteries allocate
eupyrene sperm strategically based on the intensity of sperm

M.J. Solensky, K.S. Oberhauser / Animal Behaviour 77 (2009) 465472

20
(a)

Number of eupyrene sperm

( 10 000)

Spermatophore mass (mg)

40
35
30
25
20
15
10
5

469

(a)
16
12
8
4
0

9 10 11 12 13 14 Virgin

0
1
2
3
4
Number of female's previous mates

(b)

Number of eupyrene sperm

( 10 000)

Number of eupyrene sperm

( 10 000)

20
16
12
8
4

1 2 3 4 5 6 7 8 9 10 11 12 13 14 Virgin

Number of apyrene sperm

( 1 000 000)

6
(c)

30
(b)
25
20
15
10
5
0

10

20
30
40
50
60
Mass (mg) of old spermatophores
stored from previous copulations

70

Figure 3. Relation between female mating history, as measured by (a) the number of
previous copulations completed by a female and (b) the total mass of old spermatophores stored by a female from previous copulations, and the number of eupyrene
sperm transferred by a male.

4
3
2
1
0

4 5 6 7 8 9 10 11 12 13 14 Virgin
Days since male's last mating

Figure 2. Relation between male time since last mating and (a) spermatophore size,
(b) eupyrene sperm and (c) apyrene sperm transferred. X axes show the number of
days since a nonvirgin males last mating. Although virgin males were not included in
the regression models, the means (SEs) of this group are shown by large shaded
circles at the right of each graph for comparison.

competition. Males appear to assess the intensity of sperm

competition using bursa copulatrix content as a measure of female
mating history, transferring more eupyrene sperm to females that
had stored larger amounts of spermatophore material from
previous males. It is possible that there is an additional effect of the
number of mates; females with three small spermatophores
contained marginally less spermatophore material than those with
one large spermatophore, but they received a statistically similar
number of eupyrene sperm.
In the observational study, it is possible that female mating
history was confounded with female quality, and that high-quality
females mated more often and received more sperm. However, in
our experimental study, females were pre-assigned to mating
history treatments, and these data support female mating history
as the cause of increased sperm transfer; we observed the same
pattern of eupyrene sperm allocation when male and female
mating histories were randomly pre-assigned as when they were
allowed to vary naturally.

There was no effect of female mating history on apyrene sperm

number or spermatophore size in the observational study, and only
a marginal effect in the experimental study, which featured only
three possible female mating histories (S, L or SSS spermatophores).
Males delivered marginally more apyrene sperm and smaller
spermatophores to SSS females. If males do allocate these ejaculate
components strategically, these experiments did not detect strong
allocation patterns.
The different spermatophore components serve different roles
in inuencing male reproductive success. Only the eupyrene
sperm are capable of fertilizing eggs, yet they are far outnumbered by apyrene sperm (e.g. Gage & Cook 1994; Wedell &
Cook 1999a, b; this study). Although the function of apyrene
sperm remains unclear, recent research has shown that apyrene
sperm storage delays female remating in the green-veined white
buttery (Pieris napi) (Cook & Wedell 1999), thus preserving the
last males sperm precedence. Spermatophore size also delays
female remating in monarch butteries (Oberhauser 1989, 1992),
and nutrients in the spermatophore are used by the female for
both somatic maintenance and egg production (Boggs & Gilbert
1979; Oberhauser 1989). Mixed paternity is common among
multiply mated female monarchs, with last-male advantage
being more common than rst-male advantage (Solensky 2003;
Solensky & Oberhauser, in press). A strong last-male fertilization
advantage should selectively favour male ejaculate allocation
strategies that delay female remating, such as increased apyrene
sperm transfer or increased spermatophore size.
There was a marginal effect of female mating history on spermatophore size in the experiment in which we manipulated female
mating history, with males transferring less material to females
that had already mated. If this effect is real, it would result in

470

M.J. Solensky, K.S. Oberhauser / Animal Behaviour 77 (2009) 465472

12

30

(b)
10

25

20
B
15
AB

Number of eupyrene sperm

( 10 000)

Mass (mg) of old spermatophores

stored from previous copulations

(a)

6
A
4

L
SSS
Female mating treatment group

2.5

L
SSS
Female mating treatment group

16

(c)

(d)
14

1.5

A
A

Spermatophore mass (mg)

Number of apyrene sperm

( 1 000 000)

2
12

A
A

10

A
8
6
4

0.5
2
0

L
SSS
Female mating treatment group

L
SSS
Female mating treatment group

Figure 4. Mean (a) mass of old spermatophores stored by females from previous copulations, (b) number of eupyrene sperm, (c) number of apyrene sperm and (d) mass of
spermatophore transferred by the experimental male to females that had previously received one small (S), one large (L) or three small (SSS) spermatophores. Within each graph,
different letters indicate signicant differences between groups (Bonferroni corrected Tukey post hoc: P < 0.017). Error bars show 1 SE.

a delay in female remating similar to that generated by a male

delivering a large spermatophore to a virgin female. Alternatively,
the pattern could be the result of increased nutrient investment
when females are virgins and a males paternity is assured, at least
until the next mating.
If the marginal effect of mating history on apyrene sperm
number is real, males invested more in matings with females that
had mated more times, perhaps because the apyrene sperm serve
to push a previous males sperm further back in the spermatheca. If
the marginal effect is not real, males may maximize apyrene sperm
transfer based only on their own mating history. This is consistent
with the cheap ller hypothesis, which proposes that males may
use the small, energetically less expensive apyrene sperm to ll
a females spermatheca and consequently delay remating
(Silberglied et al. 1984; Cook & Wedell 1999).
Theoretical models predict that males should increase sperm
expenditure with increasing risk of sperm competition (Parker
et al. 1997; Engqvist & Reinhold 2006, 2007). Early models predicted a decrease in sperm expenditure with increasing intensity
(i.e. number of a females prior mates) of sperm competition

(Parker et al. 1996). However, more recent models have shown

that the reverse can occur (i.e. increased sperm investment when
sperm competition intensity is high) when males have large
sperm reserves (Engqvist & Reinhold 2007), or when remating
rates are high and last-male sperm precedence is prevalent
(Engqvist & Reinhold 2006). Our results support the predictions of
Engqvist & Reinhold (2006). Males allocated more eupyrene
sperm to females when the intensity of sperm competition was
higher, as predicted for species like monarchs, in which females
mate multiple times (Hill et al. 1976; Brower et al. 1977; Oberhauser 1989; Frey 1999) and last-male sperm precedence is
common (Solensky 2003).
Males of other species use a variety of cues to assess the risk or
intensity of sperm competition, including pheromones (Siva-Jothy
& Stutt 2003; Carazo et al. 2004), the presence of rival males (Evans
et al. 2003; Zbinden et al. 2003; Pound & Gage 2004; Ramm &
Stockley 2007), female mating history (Cook & Gage 1995; Wedell
1998; Wedell & Cook 1999a), or characteristics of rival ejaculates
(Cook & Gage 1995; Wedell & Cook 1999a; Engqvist 2007). The
coercive mating system of monarchs, accompanied by an apparent

M.J. Solensky, K.S. Oberhauser / Animal Behaviour 77 (2009) 465472

loss of pheromone use (Pliske 1975), may preclude males from

using pheromones to assess sperm competition risk. Because
overwintering monarchs mate at densities that are many orders of
magnitude higher than summer monarchs (Solensky 2004), and
because most of the population is in reproductive diapause
throughout much of the overwintering period (Herman et al. 1989),
the presence of other males may be an unreliable measure of the
risk or intensity of sperm competition. Our study suggests that
male monarchs use bursa copulatrix content to assess female
mating history and subsequent risk of sperm competition. Previous
studies (Cook & Gage 1995; Wedell & Cook 1999a; Engqvist 2007)
also reported patterns of increased sperm investment by male
Lepidoptera under high-intensity sperm competition (i.e. more
sperm to females that had stored more or larger ejaculates), which
is consistent with our ndings. In contrast, Engqvist (2007) found
that resource-limited male scorpionies decreased investment
under high-intensity sperm competition (i.e. fewer sperm to
females with more sperm stored from a previous mating). One
explanation offered for these opposite patterns was the difference
in sperm precedence patterns (fair rafe in scorpionies versus
last-male precedence in the two Lepidoptera; Engqvist 2007),
which is consistent with this study, as monarch butteries also
show last-male sperm precedence (Solensky 2003; Solensky &
Oberhauser, in press).
We found no evidence of behavioural inuences on ejaculate
transfer. Although female resistance behaviours were too infrequent to be statistically analysed, we found that attempt duration
(another behavioural correlate of female mating history) had no
effect on eupyrene sperm transfer. Frey (1999) reported a correlation between female resistance behaviours and bursa copulatrix
content, but Solensky (2004) found that monarch mating behaviour and attempt duration were strongly inuenced by the initial
starting positions and the vegetation on which an attempt
occurred, neither of which correlate with female mating history.
This nding suggests that mating attempt behavioural cues may
not signal female mating history as reliably as the physiological cue
of bursa copulatrix content.
Males in many taxa strategically allocate ejaculates, using
a variety of cues to assess sperm competition (Birkhead & Mller
1998; Wedell et al. 2002; Pizzari et al. 2003; Zbinden et al. 2003;
delBarco-Trillo & Ferkin 2004; Ramm & Stockley 2007; Thomas &
Simmons 2007). Here, we show that male monarch butteries
assess the relative intensity of sperm competition based on the
amount ejaculate material stored by a female from previous
matings, and tailor their ejaculates as predicted by sperm competition models to maximize lifetime reproductive success.
Acknowledgments
We thank Lynette Batte, Anja Brunet-Rossinni, Jennifer Brophy,
Reba Batalden, Bruce Leventhal, Sara Brinda, Amy Alstad, Leah
Alstad, Andy Regan, Kevin Spragg, Beth DeLong, Amy Smith and
Evan Slanczka for their assistance in rearing and observing monarchs and counting sperm. Andy Davis provided valuable advice on
digitally measuring adult buttery size. This work was supported
by Monarchs in the Classroom and the Life Science Summer
Undergraduate Research Program at the University of Minnesota
and by the Sophomore Research Program and Howard Hughes
Medical Institute Summer Research Scholar Program at The College
of Wooster.
References
Altizer, S. M., Oberhauser, K. S. & Brower, L. P. 2000. Associations between host
migration and the prevalence of a protozoan parasite in natural populations of
adult monarch butteries. Ecological Entomology, 25, 125139.

471

Andersson, J., Borg-Karlson, A.-K. & Wiklund, C. 2000. Sexual cooperation and
conict in butteries: a male-transferred anti-aphrodisiac reduces harassment
of recently mated females. Proceedings of the Royal Society of London, Series B,
267, 12711275.
Andersson, J., Borg-Karlson, A.-K. & Wiklund, C. 2003. Antiaphrodisiacs in pierid
butteries: a theme with variation!. Journal of Chemical Ecology, 29, 14891499.
Ball, M. A. & Parker, G. A. 2007. Sperm competition games: the risk model can
generate higher sperm allocation to virgin females. Journal of Evolutionary
Biology, 20, 767779.
Bebas, P., Cymborowski, B. & Giebultowicz, J. M. 2001. Circadian rhythm of sperm
release in males of the cotton leafworm, Spodoptera littoralis: in vivo and
in vitro studies. Journal of Insect Physiology, 47, 859866.
Birkhead, T. R. & Mller, A. P. 1998. Sperm Competition and Sexual Selection. San
Diego: Academic Press.
Boggs, C. L. & Gilbert, L. E. 1979. Male contribution to egg production in butteries:
evidence for transfer of nutrients at mating. Science, 206, 8384.
Brower, L. P., Calvert, W. H., Hedrick, L. E. & Christian, J. 1977. Biological observations on an overwintering colony of monarch butteries (Danaus plexippus,
Danaidae) in Mexico. Journal of the Lepidopterists Society, 31, 232242.
Carazo, P., Sanchez, E., Font, E. & Deslis, E. 2004. Chemosensory cues allow male
Tenebrio molitor beetles to assess the reproductive status of potential mates.
Animal Behaviour, 68, 123129.
Cook, P. A. & Gage, M. J. G. 1995. Effects of risks of sperm competition on the
numbers of eupyrene and apyrene sperm ejaculated by the moth Plodia
interpunctella (Lepidoptera: Pyralidae). Behavioral Ecology and Sociobiology,
36, 261268.
Cook, P. A. & Wedell, N. 1996. Ejaculate dynamics in butteries: a strategy for
maximizing fertilization success? Proceedings of the Royal Society of London,
Series B, 263, 10471051.
Cook, P. A. & Wedell, N. 1999. Non-fertile sperm delay female remating. Nature,
397, 486.
delBarco-Trillo, J. & Ferkin, M. H. 2004. Male mammals respond to a risk of sperm
competition conveyed by odours of conspecic males. Nature, 431, 446449.
Dewsbury, D. A. 1982. Ejaculate cost and male choice. American Naturalist, 119,
601610.
Eberhard, W. G. 1996. Female Control: Sexual Selection by Cryptic Female Choice.
Princeton, New Jersey: Princeton University Press.
Engqvist, L. 2007. Male scorpionies assess the amount of rival sperm transferred
by females previous mates. Evolution, 61, 14891494.
Engqvist, L. & Reinhold, K. 2006. Theoretical inuence of female mating status and
remating propensity on male sperm allocation patterns. Journal of Evolutionary
Biology, 19, 14481458.
Engqvist, L. & Reinhold, K. 2007. Sperm competition games: optimal sperm
allocation in response to the size of competing ejaculates. Proceedings of the
Royal Society of London, Series B, 274, 209217.
Evans, J. P., Pierotti, M. & Pilastro, A. 2003. Male mating behavior and ejaculate
expenditure under sperm competition risk in the eastern mosquitosh.
Behavioral Ecology, 14, 268273.
Frey, D. 1999. Resistance to mating by female monarch butteries. In: Proceedings of
the 1997 North American Conference on the Monarch Buttery (Ed. by J. Hoth,
L. Merino, K. S. Oberhauser, I. Pisanty, S. Price & T. Wilkinson), pp. 7987.
Montreal, Canada: Commission for Environmental Cooperation.
Gage, M. J. G. 1991. Risk of sperm competition directly affects ejaculate size in the
Mediterranean fruit y. Animal Behaviour, 42, 10361037.
Gage, M. J. G. 2003. Evolutionary biology: scramble for the eggs. Nature, 426, 2223.
Gage, M. J. G. & Cook, P. A. 1994. Sperm size or numbers? Effects of nutritional
stress upon eupyrene and apyrene sperm production strategies in the
moth Plodia interpunctella (Lepidoptera: Pyralidae). Functional Ecology, 8,
594599.
Gage, M. J. G. & Morrow, E. H. 2003. Experimental evidence for the evolution of
numerous, tiny sperm via sperm competition. Current Biology, 13, 754757.
Garca-Gonzalez, F. & Gomendio, M. 2004. Adjustment of copula duration and
ejaculate size according to the risk of sperm competition in the golden egg bug
(Phyllomorpha laciniata). Behavioral Ecology, 15, 2330.
Grillet, M., Dartevelle, L. & Ferveur, J.-F. 2006. A Drosophila male pheromone
affects female sexual receptivity. Proceedings of the Royal Society of London,
Series B, 273, 315323.
Herman, W. S., Brower, L. P. & Calvert, W. H. 1989. Reproductive tract development in monarch butteries overwintering in California, USA, and Mexico.
Journal of the Lepidopterists Society, 43, 5058.
Hill, H. F., Jr, Wenner, A. M. & Wells, P. H. 1976. Reproductive behavior in an
overwintering aggregation of monarch butteries. American Midland Naturalist,
95, 1019.
Oberhauser, K. S. 1988. Male monarch buttery spermatophore mass and mating
strategies. Animal Behaviour, 36, 13841388.
Oberhauser, K. S. 1989. Effects of spermatophores on male and female
monarch buttery reproductive success. Behavioral Ecology and Sociobiology, 25, 237246.
Oberhauser, K. S. 1992. Rate of ejaculate breakdown and intermating intervals in
monarch butteries. Behavioral Ecology and Sociobiology, 31, 367373.
Oberhauser, K. S. & Frey, D. 1999. Coercive mating by overwintering male monarch
butteries. In: Proceedings of the North American Conference on the Monarch
Buttery (Ed. by W. A. Haber, L. Merino, K. Oberhauser, I. Pisanty & T. Wilkinson),
pp. 6778. Montreal, Quebec: Commission for Environmental Cooperation.
Parker, G. A. 1970. Sperm competition and its evolutionary consequences in the
insects. Biological Reviews, 45, 525567.

472

M.J. Solensky, K.S. Oberhauser / Animal Behaviour 77 (2009) 465472

Parker, G. A. 1990. Sperm competition games: rafes and roles. Proceedings of the
Royal Society of London, Series B, 242, 120126.
Parker, G. A., Ball, M. A., Stockley, P. & Gage, M. J. G. 1996. Sperm competition
games: individual assessment of sperm competition intensity by group
spawners. Proceedings of the Royal Society of London, Series B, 263, 12911297.
Parker, G. A., Ball, M. A., Stockley, P. & Gage, M. J. G. 1997. Sperm competition
games: a prospective analysis of risk assessment. Proceedings of the Royal
Society of London, Series B, 264, 17931802.
Pizzari, T., Cornwallis, C. K., Levlie, H., Jakobsson, S. & Birkhead, T. R. 2003.
Sophisticated sperm allocation in male fowl. Nature, 426, 7074.
Pliske, T. E. 1975. Courtship behavior of the monarch buttery, Danaus plexippus L.
Annals of the Entomological Society of America, 68, 143151.
Pound, N. & Gage, M. J. G. 2004. Prudent sperm allocation in Norway rats, Rattus
norvegicus: a mammalian model of adaptive ejaculate adjustment. Animal
Behaviour, 68, 819823.
Ramm, S. A. & Stockley, P. 2007. Ejaculate allocation under varying sperm
competition risk in the house mouse, Mus musculus domesticus. Behavioral
Ecology, 18, 491495.
Seth, R. K., Rao, D. K. & Reynolds, S. E. 2002. Movement of spermatozoa in the
reproductive tract of adult male Spodoptera litura: daily rhythm of sperm
descent and the effect of light regime on male reproduction. Journal of Insect
Physiology, 48, 119131.
Silberglied, R. E., Shepherd, J. G. & Dickinson, J. L. 1984. Eunuchs: the role of
apyrene sperm in Lepidoptera. American Naturalist, 123, 255265.
Siva-Jothy, M. T. & Stutt, A. D. 2003. A matter of taste: direct detection of female
mating status in the bedbug. Proceedings of the Royal Society of London, Series B,
270, 649652.
Sokal, R. R. & Rohlf, F. J. 1995. Biometry, 3rd edn. New York: W. H. Freeman.

Solensky, M. J. 2003. Reproductive tness in monarch butteries, Danaus plexippus.

Ph.D., University of Minnesota.
Solensky, M. J. 2004. The effect of behavior and ecology on male mating success in
overwintering monarch butteries (Danaus plexippus). Journal of Insect Behavior,
17, 723743.
Solensky, M. J. & Oberhauser, K. S. In press. Sperm precedence in monarch
butteries (Danaus plexippus). Behavioral Ecology.
Svard, L. & Wiklund, C. 1988. Prolonged mating in the monarch buttery Danaus
plexippus and nightfall as a cue for sperm transfer. Oikos, 51, 351354.
Thomas, M. L. & Simmons, L. W. 2007. Male crickets adjust the viability of their
sperm in response to female mating status. American Naturalist, 170, 190195.
r Zellforschung, 101,
Virkki, N. 1969. Sperm bundles and phylogenesis. Zeitschrift fu
1327.
Wedell, N. 1998. Sperm protection and mate assessment in the bushcricket
Coptaspis sp. 2. Animal Behaviour, 56, 357363.
Wedell, N. & Cook, P. A. 1999a. Butteries tailor their ejaculate in response to
sperm competition risk and intensity. Proceedings of the Royal Society of London,
Series B, 266, 10331039.
Wedell, N. & Cook, P. A. 1999b. Strategic sperm allocation in the small white
buttery Pieris rapae (Lepidoptera: Pieridae). Functional Ecology, 13, 8593.
Wedell, N., Gage, M. J. G. & Parker, G. A. 2002. Sperm competition, male prudence
and sperm-limited females. Trends in Ecology and Evolution, 17, 313320.
Yamane, T. & Miyatake, T. 2005. Intra-specic variation in strategic ejaculation
according to level of polyandry in Callosobruchus chinensis. Journal of Insect
Physiology, 51, 12401243.
Zbinden, M., Mazzi, D., Kunzler, R., Largiade`r, C. R. & Bakker, T. C. M. 2003.
Courting virtual rivals increase ejaculate size in sticklebacks (Gasterosteus
aculeatus). Behavioral Ecology and Sociobiology, 54, 205209.