Male mating behavior and costs of sexual harassment
for females in cavernicolous and extremophile
populations of Atlantic mollies (Poecilia mexicana)
Martin Plath1,2,3,4)
(1 Unit of Animal Ecology, Department of Biochemistry and Biology, University of
Potsdam, Maulbeerallee 1, 14469 Potsdam, Germany; 2 Unit of Evolutionary Biology and
Systematic Zoology, Department of Biochemistry and Biology, University of Potsdam,
Karl-Liebknecht-Str. 24-25, 14476 Potsdam, Germany; 3 Department of Zoology, University
of Oklahoma, 730 Van Vleet Oval, Norman, OK 73019, USA)
(Accepted: 7 October 2007)
Summary
The Atlantic molly Poecilia mexicana inhabits a variety of different habitat types, some of
which are characterized by physicochemical stressors, such as toxic hydrogen sulfide (H2 S),
or the fish occur in lightless subterranean (cave) habitats. Using five different populations
from surface habitats with or without H2 S, and from a sulfidic and a non-sulfidic cave, I asked
how extreme environmental conditions affect behavioral evolution in this species. I examined
male sexual behavior, and potential costs of male sexual harassment to females, determined
as the reduction of female feeding times in the presence of a harassing male. In all populations with at least one physicochemical stressor present (H2 S or absence of light), a reduction
in male sexual activity was recorded. While females from habitats without physicochemical stressors spent less time feeding around males, no such costs of male harassment were
detected in populations from extreme habitats. It is discussed that energy limitation may be
the driving force selecting for reduced male sexual activity, since fish from all habitats with
presence of physicochemical stressors show a low body condition factor under natural conditions.
Keywords: cavefish, coercive mating, hydrogen sulfide, mating tactics, sexual conflict.
4)
Author’s e-mail address: [email protected]
© Koninklijke Brill NV, Leiden, 2008
Behaviour 145, 73-98
Also available online - www.brill.nl/beh
74
Plath
Introduction
Differential interests of the sexes often lead to sexual conflicts and usually
the optimal number of copulations is lower for females than for males (Arnqvist & Rowe, 2005). Males may rely on coercion to inseminate uncooperative females (e.g., Krupa et al., 1990; Smuts & Smuts, 1993; Clutton-Brock
& Parker, 1995; McLain & Pratt, 1999; Shine et al., 2000; Fox, 2002). High
male sexual activity may come with a cost for females, because females need
to avoid male mating attempts. Male behavior that leads to changes in female
behavior can be defined as sexual harassment (Rowe et al., 1994; CluttonBrook & Parker, 1995; Reale et al., 1996; Persaud & Galef, 2003).
Livebearing fishes (Poeciliidae) are good models to study sexual harassment (Magurran & Seghers, 1994a,b; Houde, 1997; Magurran, 2001; Brewster & Houde, 2003). In poeciliids fertilization is internal, and males use
their transformed anal fin, the gonopodium, to transfer sperm (Greven, 2005).
Poeciliid females have a roughly monthly sexual cycle (Parzefall, 1973), and
they can store sperm and, thus, require few copulations to ensure complete
fertilization of several monthly broods (Constantz, 1984, 1989). Males, by
contrast, almost constantly engage in sexual behavior (e.g., guppy, Poecilia
reticulata: Magurran & Seghers, 1994a; Houde, 1997; Magurran, 2001; Atlantic molly, P. mexicana: Plath et al., 2003, 2007a; mosquitofish, Gambusia
holbrooki: Bisazza & Marin, 1995). For example, guppy females are receptive for less than 5% of the time in their reproductive cycle just after parturition (Houde, 1997; see also Farr & Travis, 1986 for sailfin molly, P. latipinna). Guppy females willingly copulate only two or three times (Kelly et al.,
1999) and typically flee from approaching males (e.g., Brewster & Houde,
2003), except when they are receptive or virgin (Liley, 1983; see also Plath
et al., 2001 and Plath & Tobler, 2007 for P. mexicana). Poeciliid males may
not only impose a cost on females by attempting forced copulations, but they
also show high degrees of sexual harassment preceding forced copulations
(e.g., nipping at the female’s gonopore, Parzefall, 1969).
Several studies were able to quantify a cost of sexual harassment as reduced feeding time for female livebearing fishes (P. reticulata: Magurran &
Seghers, 1994a; Griffiths, 1996; P. latipinna: Schlupp et al., 2001; P. mexicana: Plath et al., 2003; Gambusia holbrooki: Pilastro et al., 2003; Girardinichthys multiradiatus (Godeidae): Valero et al., 2005). Access to food is
directly correlated with growth and fecundity in poeciliid females (Hester,
Sexual harassment in extremophile mollies
75
1964; Reznik, 1983; Reznik & Miles, 1989), thus feeding time reduction is a
good indicator of a cost of sexual harassment for females. Indeed, the direct
consequences of male harassment for poeciliid females include a reduction
in the number of offspring per brood (P. reticulata: Ojanguren & Magurran,
2007; but see also Smith & Sargent, 2006 and Smith, 2007 for effects of
female competition in Gambusia affinis).
In another study, Plath et al. (2007a) examined the costs of male sexual
harassment for females in nine species of livebearing fishes (Poecilia latipinna, P. mexicana, P. salvatoris, P. reticulata, Xiphophorus cortezi, X. variatus, Gambusia affinis, G. geiseri, Heterandria formosa). Little variation was
found among species or among populations within species, suggesting that
qualitative differences in the kind of mating behavior employed by males —
such as presence or absence of courtship — have little effect on the reduction of female feeding time (Plath et al., 2007a). The only exception from this
pattern was found in the case of the cave molly, a subterranean form of the
Atlantic molly (P. mexicana) from a sulfidic cave near Tapijulapa in Tabasco,
southern Mexico (Cueva del Azufre, a.k.a. Cueva Villa Luz or Cueva de las
Sardinas; Gordon & Rosen, 1962; Walters & Walters, 1965). Cave molly
males showed little sexual behavior, and females did not feed less in the
presence of males (Plath et al., 2003, 2004a, 2007a).
A crucial aspect of the cave molly’s ecology is that the Cueva del Azufre
is not only lightless but also contains toxic levels of hydrogen sulfide (H2 S;
Langecker et al., 1996; Tobler et al., 2006). Due to the toxicity of H2 S,
the Cueva del Azufre is considered an extreme habitat for the inhabiting
P. mexicana (Tobler et al., 2006). The ecosystem of the Cueva del Azufre
appears to be based on bacterial chemoautotrophic primary production and
bat guano (Langecker et al., 1996; Parzefall, 2001). It supports a large molly
population and, thus, has been classified as energy-rich (Langecker et al.,
1996). Nevertheless, cave mollies show a very low body condition (Plath et
al., 2005; Tobler et al., 2006), and their survival under sulfidic and hypoxic
conditions is critically dependent on supply of energy-rich food (Plath et al.,
2007e). This suggests that the detoxification of H2 S is energy consuming.
Also, cave mollies need to dedicate considerable time to aquatic surface
respiration (Plath et al., 2007e), an energy- and time-consuming behavior
that detracts from the time available for feeding (Weber & Kramer, 1983).
Generally, organisms need to balance the energy allotment required for
basic somatic needs and those required for reproductive efforts (Bell &
76
Plath
Koufopanou, 1986; Stearns, 1989), and this trade-off probably differs vastly
across populations of P. mexicana. Mollies inhabiting the Cueva del Azufre
likely must invest relatively more energy in somatic maintenance than those
residing in less harsh habitats (Franssen et al., 2007; see also Sibly & Calow,
1989; Celentano & Defeo, 2006). A likely explanation for reduced male sexual behavior and absence of sexual harassment would be that over evolutionary times, extreme environmental conditions have selected against high
levels of (energy-demanding) male sexual behavior in the Cueva del Azufre
cavefish. This hypothesis would be supported if other populations inhabiting
habitats with presence of physicochemical stressors would also exhibit reduced male sexual activity and, thus, females from such populations would
also not suffer from male harassment (i.e., they would not experience a feeding time reduction).
Until recently, only one cave form of P. mexicana was known (Gordon &
Rosen, 1962; Parzefall, 1993, 2001; Proudlove, 2006), impeding a comparative analysis of male sexual behavior in P. mexicana populations that have
independently colonized caves. Recently, Pisarowicz (2005) reported the discovery of a new cave in the vicinity of the Cueva del Azufre (the Cueva Luna
Azufre), which is also inhabited by a molly population. Despite its name, the
Luna Azufre cave lacks significant amounts of hydrogen sulfide (H2 S; Tobler
et al., 2007a). Fish from the Luna Azufre cave show equally low condition
factors, suggesting that food availability is low in this cave, probably because chemoautotrophic primary production is lacking (Tobler et al., 2007a).
Mollies from both caves share characteristics like reduced (albeit functional)
eyes and pigmentation, but overall they are morphologically distinct (Tobler
et al., 2007a, unpubl. data). Indeed, an analysis of cytochrome b gene sequences indicates that the two cave forms have evolved independently, and a
microsatellite analysis suggests that there is currently no gene flow between
them (M. Plath et al., unpubl. data).
In the present study, I compared male sexual activity and the cost of male
harassment as reduced feeding time among five populations of P. mexicana,
while including the newly discovered cave population from the non-sulfidic
Luna Azufre cave. I employed populations that inhabit habitats with different
combinations of the environmental factors light/darkness and sulfidic/nonsulfidic, and I asked if the described behavioral reduction processes occur in
all habitats in which environmental stressors (absence of light and/or presence of toxic hydrogen sulfide) act as selective forces.
Sexual harassment in extremophile mollies
77
Materials and methods
Study system
Poecilia mexicana is widespread in freshwater surface habitats along the Atlantic versant of Central America (Miller, 2005). The Cueva del Azufre is
about 300 m deep and has several breaks in the ceiling in the front parts
(Gordon & Rosen, 1962; Hose & Pisarowicz, 1999). A cave creek flows
through the cave, forming several shallow pools that are partially divided by
riffle passages. While the front cave chambers receive some dim light, the
inner parts of the cave are lightless, and the molly population from the innermost cave chamber XIII constantly lives in the dark. Pronounced genetic
differentiation between the fish from the Cueva del Azufre, El Azufre and the
nearby Arroyo Cristal (a tributary to the Río Oxolotan) was detected using
microsatellite analysis, suggesting local adaptation and restricted gene flow
among populations (Plath et al., 2007d).
The Luna Azufre is a considerably smaller cave (Tobler et al., 2007a).
Both caves are located within different hills that are separated by a valley
and a creek. Interestingly, very large males do not occur in this cave population, such that males are, on average, smaller than those from the Cueva del
Azufre (R. Riesch et al., unpubl. data). Accordingly, Luna Azufre males from
stock populations were also smaller than males from the other populations
(Table 1).
Another aspect of the mating system of P. mexicana I considered in this
study was size-dependent male sexual activity. Large poeciliid males are typically favored by females (e.g., P. latipinna: MacLaren & Rowland, 2006;
MacLaren, 2007; P. mexicana: Plath et al., 2004b, 2007b), and are superior in dyadic conflicts (e.g., green swordtails, Xiphophorus hellerii: Franck
& Ribowski, 1989; Ribowski & Franck, 1993). Consequently, both female
choice and male competition select for large male body size. Small males
may compensate for this by using alternative mating tactics (Farr, 1989).
Moreover, small males may just show more sexual behavior on encounter
of a female than large males (P. latipinna: Schlupp et al., 2001; P. mexicana: Plath et al., 2003). Indeed, in surface dwelling Atlantic mollies, small
males exhibit ‘ambushing’ behavior, hiding near groups of females, while relying on forced copulations in the absence of large males (Parzefall, 1969).
Even large P. mexicana males do not show the elaborate courtship known,
34.19 ± 0.47
36.46 ± 1.34
32.30 ± 0.61
34.13 ± 0.58
32.50 ± 1.21
33.72 ± 0.66
35.75 ± 1.30
31.45 ± 0.52
32.91 ± 0.58
29.58 ± 1.18
Focal female Partner female
SL (mm)
SL (mm)
28.80 ± 0.60
30.79 ± 1.00
27.48 ± 0.60
29.89 ± 0.57
24.26 ± 0.30
Male SL
(mm)
35.66 ± 5.20
26.08 ± 5.65
10.03 ± 2.59
13.83 ± 2.48
7.32 ± 2.29
10.39 ± 1.89
5.54 ± 2.11
3.13 ± 0.99
3.35 ± 0.77
3.09 ± 0.51
47.86 ± 7.09
42.56 ± 4.28
17.43 ± 5.51
15.08 ± 3.99
9.22 ± 13.20
Pre-copulatory
Copulatory attempts
Sum of mating behaviors
behavior (nipping) (gonopodial thrusts) (estimated marginal means1 )
appearing in the model are evaluated at the following values: male body size = 28.40 mm, female size = 33.71 mm.
56
26
40
71
37
Tampico
Río Oxolotan
El Azufre
Cueva del Azufre
Luna Azufre
1 Covariates
N
Population
Table 1. Standard lengths (SL) of the test fish and counts of male sexual behavior (mean ± SE). Estimated marginal
means for male sexual activity (sum of male mating behaviors) are estimated from a generalized linear model (see
main text).
78
Plath
Sexual harassment in extremophile mollies
79
for example, from sailfin mollies (Parzefall, 1969; Plath et al., 2007a). Sizedependent rates of mating behavior were found to be absent in the cave molly
(Cueva del Azufre), and in some studies even a slightly positive correlation
between male size and sexual activity was found (Plath et al., 2003, 2004a,c;
Riesch et al., 2006a). Hence, the relationship between male body size and
male sexual activity was also analyzed in the present study.
Origin and maintenance of study organisms
Randomly out-bred laboratory stocks of five populations of P. mexicana were
used. One population was collected from the Río Oxolotan, a river with
mostly clear water in the vicinity of the caves (Tobler et al., 2006). A second
population from a non-sulfidic surface habitat was collected in the brackish
coastal waters near Tampico (Tamaulipas, eastern Mexico). Another population was collected from just outside of the exit of the Cueva del Azufre.
It occurs in the milky, sulfidic water of the El Azufre, which is flowing out
of the cave (Tobler et al., 2006). The fourth population was collected from
the rearmost cave chamber XIII of the Cueva del Azufre (after Gordon &
Rosen, 1962). Finally, the fish from the newly discovered Luna Azufre were
collected. In total, N = 75 new trials were conducted for this study (N = 37
of the newly discovered Luna Azufre cavefish, N = 32 Tampico and N = 6
Río Oxolotan). I included N = 155 trials that were re-analyzed from previous studies using the same experimental design: Plath et al. (2003), N = 20
(Río Oxolotan), N = 25 (Cueva del Azufre) and N = 20 (El Azufre); Plath
et al. (2004a), N = 20 (Cueva del Azufre); Plath et al. (2006a), N = 20
(El Azufre); Plath et al. (2007a), N = 24 (Tampico), N = 26 (Cueva del
Azufre).
Individuals examined in the course of this study were reared in large (ca.
1000 l) tanks in a greenhouse of the University of Oklahoma in Norman.
The tanks contained naturally growing algae as well as a variety of naturally
occurring invertebrates like chironomid larvae, copepods and amphipods, on
which the fish could feed. In addition, the fish were supplied with flake food
every two days. All fish used were sexually mature and had interacted with
the opposite sex; thus, all females were most likely pregnant. Where data
were re-analyzed from previous studies, fish had been kept in the laboratory
of the Zoological Institute of the University of Hamburg in 100–200 l tanks
at 25–30◦ C under a 16:8 h light/dark cycle. Note that also the cavefish were
maintained at light.
80
Plath
The stocks had been maintained in the laboratory for different periods of
time: the Tampico population was collected in 1995 and was tested between
2000 and 2003, as well as in 2006. The Río Oxolotan population was collected in 1995 and 1998, and the populations from the Cueva del Azufre
and El Azufre were first collected in 1970 and repeatedly refreshed in 1975,
1982 and 1996, and data for those populations were collected between 2000
and 2003 (Plath et al., 2003, 2004a, 2006a). Another stock of the Cueva del
Azufre population (Plath et al., 2007a), as well as the Luna Azufre population originated from animals collected in 2004 and 2006; data were collected
between 2006 and 2007. It could be argued that prolonged periods of captivity could have lead to decreased sexual activity, i.e., if inbreeding plays
a role. However, in the case of the two cave populations, animals from the
second to third laboratory generation were available, while the Río Oxolotan
and Tampico males all came from older laboratory stocks; nevertheless, more
sexual behavior was seen in the Río Oxolotan and the Tampico population
than in the cave populations (see below). In the case of the Cueva del Azufre
cave population, data were suitable for a comparison of male sexual activity
between stocks that had been maintained in captivity for a longer (Hamburg) or shorter period of time (Norman). Male sexual activity (dependent
variable) was compared using GLM, whereby male body size and female
body size were treated as covariates, and ‘laboratory’ was used as a between
subjects factor. However, no significant effects were observed (F < 0.10,
p > 0.37).
All fish were acclimated to laboratory conditions for 24 h and were fed
food tablets, which guaranteed that the fish would be habituated to, and feed
on food tablets during the tests. Then, males were isolated from females in
small, visually separated aquaria for another 24 h, after which period P. mexicana males readily show mating behavior (Schlupp & Plath, 2005). Meanwhile the focal females were isolated in small groups (4–5 individuals) in 50
l aquaria and were not fed for 24 h before the tests, such that focal females
were motivated to feed throughout the test. Males and partner females were
fed ad libitum in the morning just prior to the tests, making sure that males
would not trade off foraging and mating (see Abrahams, 1993).
Feeding tests
Test tanks were used in alternating order. Each of the tanks (50×30×25 cm,
length × height × width) was filled to three-fourths with aged tap water.
Sexual harassment in extremophile mollies
81
The base was covered with a thin layer of white gravel. Water temperature
was maintained at 24–26◦ C with the aid of an aquarium heater. Illumination
was provided by four 60 W lamps on the ceiling of the test room. Prior to
a feeding test, a food tablet (TetraMin tropical tablets) was placed on the
bottom of the test tank, centrally near the front wall. A transparent Plexiglas
box (30 × 8 × 8 cm, open at the top and base) was placed in the back
center of the tank to hold the focal female during the acclimation phase.
To initiate a trial, a focal female was introduced into the acclimation box,
and a male or a female partner fish was introduced into the test tank. Both
fish were given 5 min to acclimatize. The focal female was then released and
the behavior observation was started. I measured the time the focal female
spent feeding on the presented stationary food source, from the surface of
the water, the aquarium walls and bottom, and on floating matter during a
five-minute observation period. Whenever a male partner was present, I also
recorded the number of sexual behaviors. Male sexual behavior was recorded
as pre-mating behavior involving body contact (i.e., nipping at the female
gonopore) and gonopodial thrusts (i.e., copulations and copulatory attempts;
Parzefall, 1969; Farr, 1989). Nipping at the female genital pore typically
precedes copulation attempts in mollies (Parzefall, 1969, 1973; Sumner et
al., 1994).
After the first part of a trial, the focal female was introduced into the
acclimation tube again, the first partner was removed, and a partner fish of
the opposite sex was introduced. Hence, each focal female experienced two
subsequent experimental situations during which she could feed either with
a male or with another female. The order of presentation (male or female
partner first) was balanced. After a trial, all fish involved were measured for
standard length to the closest millimeter (Table 1).
Statistical analyses
Overall differences in male mating activity (the sum of all sexual behaviors)
across populations were tested using a generalized linear model (GLM),
whereby ‘population’ was treated as a between subjects factor. To test for
an effect of male body size on male mating activity (e.g., Schlupp et al.,
2001), male standard length was included as a covariate. To test for an effect
of female body size (i.e., male mating preferences for large females, e.g.,
Plath et al., 2006b, 2007c) the focal females’ standard lengths were included
as another covariate.
82
Plath
One of the major questions of this study was whether females would spend
less time feeding in the presence of a male and whether this feeding time
reduction would differ among populations. In one approach, I compared
female feeding times (with a male or with a female) within each of the
different populations using paired t-tests.
In another approach, female feeding time reduction (determined as the
relative time spent feeding in the presence of a male) was compared among
populations. The relative time spent feeding with a given male was calculated as follows: (time spent feeding with that male/(time spent feeding with
the respective partner female + time spent feeding with the male)). Hence,
values < 0.50 would indicate that the focal females fed less with the male
partner, 0.50 would indicate no change at all and values > 0.50 would indicate that the females fed more with the male partner. All relative data were
arcsine (square-root)-transformed for the analyses. Again, I employed GLM,
in which ‘population’ was a between subjects factor. To account for differences in males’ readiness to mate, male mating activity (individual values
for the sum of all sexual behaviors) was included as a covariate. Because
the overall feeding motivation of females might also influence the degree to
which the females suffer from male harassment in terms of a feeding time
reduction, the total feeding time of the focal females (feeding time with female + with male partner) was included as another covariate. Where interactions with one of the covariates were significant, standardized residuals from
GLM were used to calculate post hoc Pearson correlations, which allowed
qualitatively comparing correlation coefficients.
Results
Male sexual behavior
Male sexual activity
In the analysis of male sexual activity (sum of all male mating behaviors) the
interaction term ‘population × male size × female size’ had no statistically
significant effect (F4,210 = 0.62, p = 0.65), such that only interactions up to
the level of two were analyzed. Overall, there was a significant difference in
male sexual behavior among populations (Table 2), and males from the two
clear water surface habitats (Río Oxolotan and Tampico) showed more sexual behavior than males from the sulfur creek and the two cave populations
83
Sexual harassment in extremophile mollies
Table 2. Generalized linear model (GLM) using male sexual activity (sum
of all male sexual behaviors per 5 min) as dependent variable.
Source
Population
Male size
Female size
Population × male size
Population × female size
Male size × female size
Error
df
Mean square
F
p
4
1
1
4
4
1
214
7799.49
2236.12
1252.59
4237.90
3333.05
1470.86
1000.37
7.80
2.24
1.25
4.24
3.33
1.47
<0.0001
0.14
0.26
0.003
0.011
0.23
(Table 1). A post hoc LSD test revealed significant differences between the
Río Oxolotan and Tampico populations and any of the other three populations (0.0001 p 0.027), while there was no difference between the Río
Oxolotan and the Tampico population (p = 0.52) or among the other three
populations (p 0.57).
Size-dependent male sexual activity
The interaction term ‘population × male size’ had a significant effect, indicating different slopes for the correlations between male body size and
the number of sexual behaviors across populations (Table 2). Indeed, negative correlations between male body size and the number of male mating
behaviors (standardized residuals) were found in the case of the two clearwater surface populations (Pearson correlations, Río Oxolotan: rP = −0.39,
p = 0.052, N = 26; Tampico: rP = −0.37, p = 0.005, N = 56; Figure 1a), indicating typical size-dependent male sexual activity in these two
populations, with smaller males showing a higher sexual activity than larger
ones. By contrast, no such correlations were found in the other three populations (Cueva del Azufre: rP = +0.07, p = 0.55, N = 71; Luna Azufre:
rP = +0.05, p = 0.77, N = 37; El Azufre: rP = −0.06, p = 0.71, N = 40;
Figure 1b).
Male mate choice for large female size
The interaction term ‘population × female size’, too, had a significant effect,
indicating different responses of the males to different-sized females across
populations (Table 2). There was a significant positive correlation between
84
Plath
Figure 1. The correlations between male sexual activity (standardized residuals) and male
body size (standard length). Residuals were obtained from a generalized linear model (see
Table 2).
Sexual harassment in extremophile mollies
85
female body size and male sexual behavior (standardized residuals) in the
Cueva del Azufre cavefish (Pearson correlation: rP = +0.31, p = 0.008,
N = 71) and in males from the sulfidic surface creek (El Azufre; rP =
+0.50, p = 0.001, N = 40), suggesting male mate choice for large female
size in these two populations (Figure 2). By contrast, I detected no significant
correlation in the surface population from Tampico (rP = −0.04, p = 0.77,
N = 56) and the newly discovered Luna Azufre cavefish (rP = −0.25,
P = 0.13, N = 37), and even a negative correlation was found in the Río
Oxolotan population (rP = −0.41, P = 0.037, N = 26; Figure 2).
Female feeding time reduction
Female feeding time in the presence of a female or a male
A pair-wise comparison of female feeding times between the part of the tests
involving a male and a female partner revealed that focal females spent less
time feeding in the presence of a male only in the two surface populations
from non-sulfidic habitats (Río Oxolotan and Tampico; Table 3).
Population comparison of female feeding time reduction
In the GLM, the interaction term ‘population × total feeding time × male
sexual behavior’ had no statistically significant effect (F4,210 = 1.32, p =
0.26), and only interactions up to a level of two were analyzed. Overall, there
was a significant difference in feeding time reduction among populations
(Table 4; Figure 3). In a post hoc comparison, the Río Oxolotan and Tampico
populations differed from the other three populations (LSD test: 0.0003 p 0.056), while there was no difference between the Río Oxolotan and
the Tampico population (p = 0.32) or among the other three populations
(p 0.41).
Effect of male sexual activity
There was also an effect of the interaction term ‘population × male sexual
activity’ (Table 4), indicating differences across populations in the extent
to which female feeding times were affected by male sexual activity. Posthoc Pearson correlations using standardized residuals revealed a little congruent pattern: there was a significant negative correlation in the Tampico
population (rP = −0.34, p = 0.010, N = 56), no correlation in the Río
86
Plath
Figure 2. The correlations between male sexual activity (standardized residuals) and female
body size (standard length). Residuals were obtained from a generalized linear model (see
Table 2).
87
Sexual harassment in extremophile mollies
Table 3. The mean (± SE) time focal females of the five populations of P.
mexicana spent feeding when interacting with another female or a male.
Population
Habitat
characteristics
Time spent
feeding with
female (s)
Time spent
feeding with
male (s)
97.54 ± 10.59
68.61 ± 9.20
Difference in
feeding times,
paired t-test
t55 = 3.80,
p < 0.00011
Río Oxolotan
Surface, non-sulfidic 40.77 ± 8.60
31.38 ± 8.41 t25 = 2.10,
p = 0.0461
El Azufre
Surface, sulfidic
74.65 ± 10.36 80.60 ± 10.15 t39 = −0.48,
p = 0.64
Cueva del Azufre Cave, sulfidic
65.54 ± 7.37
64.55 ± 7.61 t70 = 0.19,
p = 0.851
Luna Azufre
Cave, non-sulfidic
103.19 ± 13.78 112.11 ± 11.48 t36 = −0.65,
p = 0.52
Tampico
1 Test
Surface, non-sulfidic
on log-transformed data.
Table 4. Generalized linear model (GLM) using the relative time females
spent feeding with a partner male (arcsine(square-root)-transformed) as dependent variable.
Source
Population
Total feeding time
Male sexual activity
Population × total feeding time
Population × male sexual activity
Total feeding time × male sexual activity
Error
df
Mean square
F
p
4
1
1
4
4
1
214
0.39
0.23
0.06
0.08
0.31
0.00
0.11
3.47
2.09
0.53
0.74
2.83
0.00
0.009
0.15
0.47
0.57
0.026
0.96
Oxolotan population (rP = +0.25, p = 0.22, N = 26), the Cueva del
Azufre (rP = −0.01, p = 0.92, N = 71) or the Luna Azufre population
(rP = +0.04, p = 0.80, N = 37), and even a tendency towards a positive
correlation in the El Azufre population (rP = +0.30, p = 0.061, N = 40;
Figure 4). The latter finding may indicate that — in this population with absence of male harassment — male sexual activity and female feeding rates
are to some extent correlated, because both indicate general activity levels.
88
Plath
Figure 3. Reduction of females’ feeding time by male harassment (% time spent feeding
with a male −0.50; mean ± SE). Negative values indicate that females spent less time feeding
in the presence of a male.
Discusssion
In all populations with at least one physicochemical stressor present (H2 S
or absence of light), a reduction in male sexual activity was observed compared with males from populations that inhabit ‘benign’ habitats. While females from habitats without physicochemical stressors spent less time feeding around males, no such costs of male harassment were detected in populations from extreme habitats.
How could the presence of physiochemical stressors have selected for low
male sexual activity in those populations? Inside the Cueva del Azufre, H2 S
concentrations can reach more than 300 µM, and also in El Azufre, toxic
concentrations of H2 S can be found (10–41 µM; Tobler et al., 2006). The
cave environment in the Luna Azufre differs from the Cueva del Azufre by
the generally much lower H2 S concentrations (0–4 µM; Tobler et al., 2007a).
These ecological differences between the caves are also reflected by the low
abundance of sulfide-oxidizing bacteria in the Luna Azufre, which cover almost all wet surfaces in the Cueva del Azufre (Pisarowicz, 2005; Tobler et
al., 2007a). Caves are usually considered to be energy-poor habitats (Poulson & Lavoie, 2000), but because of the presence of bacterial primary production and the input of bat guano, the Cueva del Azufre was suggested to
be energy-rich even compared to surface habitats (Langecker et al., 1996;
Sexual harassment in extremophile mollies
89
Figure 4. The correlations between the reduction of female feeding time in the presence of
a male (determined as the relative time spent feeding around a male) and male sexual activity
(sum of all sexual behaviors). Standardized residuals were obtained from a generalized linear
model (see Table 4).
90
Plath
Parzefall, 2001). Bats and accumulations of bat guano can also be found in
the Luna Azufre and seem to provide the energy basis for the cave ecosystem
(M. Tobler, unpubl. data). Although the Cueva del Azufre is considered to be
an energy-rich habitat, the mollies inhabiting the cave appear to be malnourished (Plath et al., 2005, Tobler et al., 2006) and energy availability affects
the short-term survival of the fish (Plath et al., 2007e). It has been hypothesized that the poor body condition is caused by costly adaptations necessary to cope with the high concentration of H2 S (Tobler et al., 2006; Plath
et al., 2007e). This, however, seems not to apply for the Luna Azufre, where
very low concentrations of H2 S were detected, but nonetheless the inhabiting
mollies show equally low condition factors (Tobler et al., 2007a). This points
to low food availability in the Luna Azufre cave, in which chemoautotrophic
primary production by sulfide-oxidizing bacteria likely plays an insignificant
role.
Consequently, it seems likely that energy-limitation in sulfidic and/or dark
habitats was the major evolutionary force selecting for a reduction of energy
demanding behaviors (such as sexual behavior) in the El Azufre, Cueva del
Azufre and Luna Azufre populations. A similar explanation is likely for the
described reduction in aggressive behavior (Parzefall, 2001). Because labreared individuals have been used in this study, and tests were conducted in
light, and in the absence of hydrogen sulfide, behavioral differences observed
here must have a genetic basis.
Costs of male harassment
Females from the two habitats without physicochemical stressors (Río Oxolotan and Tampico) spent less time feeding in the presence of a male than
in the company of another female, but no such effect of male harassment
on females’ feeding rates was detected in the other three populations with
reduced male sexual activity.
When sailfin molly (P. latipinna) females were allowed to feed in the presence of a male that could only visually, but not physically interact with the
female, focal females nonetheless spent considerably less time feeding than
in the presence of a female (Plath et al., 2007a). This indicates that besides
a direct effect of male sexual activity on female feeding rates (as attested
for example by the negative correlation between female feeding times and
male sexual activity in the Tampico population), general shift in females’
Sexual harassment in extremophile mollies
91
time allocations due to increased vigilance play an important role (Plath et
al., 2007a). In the populations with presence of physicochemical stressors,
male sexual activity is reduced (see above), and selection on females to avoid
males is probably very low.
Generally, it remains puzzling how large molly populations occur in the
caves despite the harshness of the environmental conditions (Parzefall, 1993;
Tobler et al., 2006). Ecological benefits of life in the Cueva del Azufre
include the lack of interspecific competition (Tobler et al., 2006), lack of
avian and piscine predators (Tobler et al., 2006; Riesch et al., 2006b; Plath &
Schlupp, 2007; but see Tobler et al., 2007b for predation by giant water bugs
(Belostoma sp.)) and reduced parasitization (Tobler et al., 2007c). I propose
that — from the female perspective — reduced male sexual harassment, as
an evolutionary consequence of life under extreme environmental conditions,
acts as another (non-ecological) benefit of life under extreme environmental
conditions. For example, guppy (P. reticulata) females produce less offspring
when reared with males (Ojanguren & Magurran, 2007), but this reduction in
female fitness will not occur in P. mexicana populations inhabiting extreme
habitats.
Absence of size-dependent male sexual activity
Although a very similar polymorphism in adult male body size is found
in surface and cave dwelling (Cueva del Azufre) P. mexicana populations
(Plath et al., 2003), no size-dependent male sexual activity was found in
the two cave and the El Azufre populations. In some species such as swordtails, genus Xiphophorus (Kallman & Borkoski, 1978; Zimmerer & Kallman,
1989; Ryan et al., 1992), a genetic basis of male size polymorphism has been
demonstrated, but little is known so far about the heredity of male size in
other species like P. mexicana. A genetic basis of male size polymorphism,
however, is not necessary for the maintenance of size-dependent male sexual
activity. It must be mentioned though that males are, on average, smaller in
the Luna Azufre population, where very large males do not occur. Hence,
size-dependent male sexual activity in this population may be non-existent,
just because there is little variation in adult male body size.
Only in the Río Oxolotan and the Tampico populations, both of which
inhabit habitats without physicochemical stressors, did small males show
increased levels of sexual activity, as attested by the negative correlation between male body size and the number of male sexual behaviors. In previous
92
Plath
studies, even positive correlations between male size and male sexual activity could be detected in Cueva del Azufre males (Plath et al., 2003, 2004c;
Riesch et al., 2006a). This, however, was not confirmed in the present study.
Increased sexual activity in small males is probably very energy demanding;
thus, it is likely that the same selection against energy demanding behaviors
(see above) is responsible for the evolutionary reduction of size-dependent
male sexual activity.
Male mate choice for large female body size
A positive correlation between female body size and male sexual activity
was found in the El Azufre and Cueva del Azufre populations, but not in
the other populations, including the newly discovered Luna Azufre cavefish.
From these results it seems that the male mating preference for large female size is stronger in the El Azufre and Cueva del Azufre populations than
in the other studied populations. Indeed, using simultaneous dichotomous
choice tests, Plath et al. (2006b) found males from the Cueva del Azufre to
exhibit a significantly stronger association preference for large females than
males from the Río Oxolotan. This was interpreted by male mate choice
being more important in the sulfidic, hypoxic and energy limited habitat,
since reproductive efforts become relatively more expensive in energy limited environments. Why no mating preference relative to the females’ size
was found in the Luna Azufre males, remains unclear. Aquatic surface respiration (see Introduction) is important only in the sulfidic habitats and limits
the time available for mating in the sulfidic and hypoxic habitats, thereby
restricting the time available for mating. It is tempting to speculate that selection to evolve this male mating preference for large female size is lower
in the non-sulfidic habitat, because the time budgets of males are not constrained by aquatic surface respiration. However, simultaneous choice tests
will be needed to determine whether this male mating preference for large
female size is indeed weaker in the Luna Azufre cavefish.
In the Río Oxolotan population, even a negative correlation between female size and the number of male sexual behaviors was detected. It seems
that larger females are more able to flee from males in this population. This
idea is congruent with the proposed arms race between the sexes, where females are expected to evolve ‘resistance traits’ in response to fast evolving
male ‘persistence traits’ (Arnqvist & Rowe, 2005). The experimental design
Sexual harassment in extremophile mollies
93
of the present study can also be viewed as a sequential choice test, allowing
a male to interact with one female at a time, whereby the observed number of male sexual behaviors is also affected by the behavior of the females.
Indeed, in simultaneous, dichotomous association preference tests, in which
two stimulus females were confined to a specific location and, thus, could not
flee, also males from the Río Oxolotan spent considerably more time near a
large than near a small female (Plath et al., 2006b).
Acknowledgements
I am indebted to J. Parzefall (Hamburg) for valuable discussions and to I. Schlupp, M. Tobler
and R. Riesch for their continuous support. M. Tobler, C. Bleidorn and A. Apio kindly read
and helped to improve a previous manuscript draft. Two anonymous reviewers provided
very helpful comments. The Mexican Government kindly issued collecting and research
permits (Permiso de pesca de fomento numbers: 291002-613-1577, DGOPA/5864/260704/2408 and DGOPA/16988/191205/-8101). A. Taebel-Hellwig, T.H. Dirks, I.D. Schmidt and
the aquarium team at the Zoological Institute and Museum in Hamburg provided technical
support. Financial support came from the DFG (Pl 470/1-1, Pl 470/1-2). The experiments
reported in this paper comply with the current laws on animal experimentation in Germany
and the USA.
References
Abrahams, M. (1993). The trade-off between foraging and courting in male guppies. —
Anim. Behav. 45: 673-681.
Arnqvist, G. & Rowe, L. (2005). Sexual conflict. — Princeton University Press, Princeton,
NJ.
Bell, G. & Koufopanou, V. (1986). The cost of reproduction. — In: Oxford surveys in evolutionary biology, Vol. 3 (Dawkins, R. & Ridley, M., eds). Oxford University Press,
Oxford, p. 83-131.
Bisazza, A. & Marin, G. (1995). Sexual selection and sexual size dimorphism in the eastern
mosquitofish Gambusia holbrooki (Pisces: Poeciliidae). — Ethol. Ecol. Evol. 7: 169183.
Brewster, J. & Houde, A.E. (2003). Are female guppies more likely to flee when approached
by two males? — J. Fish. Biol. 63: 1056-1059.
Celentano, E. & Defeo, O. (2006). Habitat harshness and morphodynamics: life history traits
of the mole crab Emerita brasiliensis in Uruguayan sandy beaches. — Mar. Biol. 149:
1453-1461.
Clutton-Brock, T.H. & Parker, G.A. (1995). Sexual coercion in animal societies. — Anim.
Behav. 49: 1345-1365.
Constantz, G.D. (1984). Sperm competition in poeciliid fishes. — In: Sperm competition and
the evolution of animal mating systems (Smith, R., ed.). Academic Press, Orlando, FL,
p. 456-485.
94
Plath
Constantz, G.D. (1989). Reproductive biology of poeciliid fishes. — In: Ecology and evolution of livebearing fishes (Poeciliidae) (Meffe, G.K. & Snelson, F.F., eds). Prentice Hall,
Englewood Cliffs, NJ.
Farr, J.A. (1989). Sexual selection and secondary sexual differentiation in Poeciliids: Determinants of male mating success and the evolution of female choice. — In: Ecology and
evolution of livebearing fishes (Poeciliidae) (Meffe, G.K. & Snelson, F.F., eds). Prentice
Hall, Englewood Cliffs, NJ, p. 91-123.
Farr, J.A. & Travis, J. (1986). Fertility advertisement by female sailfin mollies, Poecilia
latipinna (Pisces: Poeciliidae). — Copeia: 467-472.
Fox, E.A. (2002). Female tactics to reduce sexual harassment in the Sumatran orangutan
(Pongo pygmaeus abelii). — Behav. Ecol. Sociobiol. 52: 93-101.
Franck, D. & Ribowski, A. (1989). Escalating fights for rank-order position between male
swordtails (Xiphophorus helleri): Effects of prior rank-order experience and information transfer. — Behav. Ecol. Sociobiol. 24: 133-143.
Franssen, C., Tobler, M., Riesch, R., García de León, F.J., Tiedemann, R., Schlupp, I. &
Plath, M. (2007). Sperm production in an extremophile fish, the cave molly (Poecilia
mexicana, Poeciliidae, Teleostei). — Aquat. Ecol. (in press).
Gordon, M.S. & Rosen, D.E. (1962). A cavernicolous form of the Poeciliid fish Poecilia
sphenops from Tabasco, México. — Copeia: 360-368.
Greven, H. (2005). Structural and behavioral traits associated with sperm transfer in Poeciliinae. — In: Viviparous fishes (Uribe, M.C. & Grier, H.J., eds). New Life Publications,
Homestead, FL, p. 145-163.
Griffiths, S.W. (1996). Sex differences in the trade-off between feeding and mating in the
guppy. — J. Fish. Biol. 48: 891-898.
Hester, F.J. (1964). Effects of food supply on fecundity in the female guppy, Lebistes reticulatus (Peters). — J. Fish. Res. Board Can. 21: 757-764.
Hose, L.D. & Pisarovicz, J.A. (1999). Cueva de Villa Luz, Tabasco, Mexico: reconnaissance
study of an active sulfur spring cave. — J. Cave Karst Stud. 61: 13-21.
Houde, A.E. (1997). Sex, color and mate choice in guppies. — Princeton University Press,
Princeton, NJ.
Kallman, K.D. & Borkoski, V. (1978). A sex-linked gene controlling the onset of sexual
maturity in female and male platyfish (Xiphophorus maculatus), fecundity in females
and adult size in males. — Genetica 89: 79-119.
Kelly, C.D., Godin, J.-G.J. & Wright, J.M. (1999). Geographic variation in multiple paternity
within natural populations of the guppy. — Proc. Roy. Soc. Lond. B Biol. 266: 24032408.
Krupa, J.J., Leopold, W.R. & Sih, A. (1990). Avoidance of male giant water striders by
females. — Behaviour 115: 247-253.
Langecker, T.G., Wilkens, H. & Parzefall, J. (1996). Studies on the trophic structure of an
energy rich Mexican cave (Cueva de las Sardinas) containing sulfurous water. — Mem.
Biospeol. 23: 121-125.
Liley, N.R. (1983). Hormones, pheromones, and reproductive behavior in fishes. — In: Fish
physiology, Vol. 3 (Hoar, W.S. & Randall, D.J., eds). Academic Press, New York, NY,
p. 73-116.
Magurran, A.E. (2001). Sexual conflict and evolution in Trinidadian guppies. — Genetica
112/113: 463-474.
Sexual harassment in extremophile mollies
95
Magurran, A.E. & Seghers, B.H. (1994a). A cost of sexual harassment in the guppy, Poecilia
reticulata. — Proc. R. Soc. Lond. – Series B Biol. 258: 89-92.
Magurran, A.E. & Seghers, B.H. (1994b). Sexual conflict as a consequence of ecology:
evidence from guppy, Poecilia reticulata, populations in Trinidad. — Proc. R. Soc.
Lond. Series B Biol. 255: 31-36.
MacLaren, R.D. (2007). The effects of male proximity, apparent size, and absolute size on
female preference in the sailfin molly, Poecilia latipinna. — Behaviour 143: 1457-1472.
MacLaren, R.D. & Rowland, W.J. (2006). Differences in female preferences for male body
size in Poecilia latipinna using simultaneous vs sequential stimulus presentation designs. — Behaviour 143: 273-292.
McLain, D.K. & Pratt, A.E. (1999). The cost of sexual coercion and heterospecific sexual harassment on the fecundity of a host-specific, seed-eating insect (Neacoryphus bicrucis).
— Behav. Ecol. Sociobiol. 46: 164-170.
Miller, R.R. (2005). Freshwater fishes of Mexico. — University of Chicago Press, Chicago,
IL.
Ojanguren, A.F. & Magurran, A.E. (2007). Male harassment reduces short-term female fitness in guppies. — Behaviour 144: 503-514.
Parzefall, J. (1969). Zur vergleichenden Ethologie verschiedener Mollienesia-Arten einschließlich einer Höhlenform von Mollienesia sphenops. — Behaviour 33: 1-37.
Parzefall, J. (1973). Attraction and sexual cycle of poeciliids. — In: Genetics and mutagenesis
of fish (Schröder, J.H., ed.). Springer, Berlin, p. 177-183.
Parzefall, J. (1993). Behavioural ecology of cave-dwelling fishes. — In: Behaviour of teleost
fishes, 2nd edn (Pitcher, T.J., ed.). Chapman and Hall, London, p. 573-608.
Parzefall, J. (2001). A review on morphological and behavioural changes in the cave molly
Poecilia mexicana from Tabasco, Mexico. — Environ. Biol. Fish. 50: 263-275.
Persaud, K.N. & Galef, B.G. (2003). Female Japanese quail aggregate to avoid sexual harassment by conspecific males: a possible cause of conspecific cueing. — Anim. Behav. 65:
89-94.
Pilastro, A., Benetton, S. & Bisazza, A. (2003). Female aggregation and male competition
reduce costs of sexual harassment in the mosquitofish Gambusia holbrooki. — Anim.
Behav. 65: 1161-1167.
Pisarowicz, J. (2005). Return to Tabasco. — Ass. Mexican Cave Stud. Newslett. 28: 27-57.
Plath, M. & Schlupp, I. (2007). Parallel evolution leads to reduced shoaling behavior in
two cave-dwelling populations of Atlantic mollies (Poecilia mexicana, Poeciliidae,
Teleostei). — Environ. Biol. Fish. (in press).
Plath, M. & Tobler, M. (2007). Sex recognition in surface- and cave-dwelling Atlantic molly
females (Poecilia mexicana, Poeciliidae, Teleostei): influence of visual and non-visual
cues. — Acta Ethol. 10: 81-88.
Plath, M., Körner, K.E., Schlupp, I. & Parzefall, J. (2001). Sex recognition and female preferences of cave mollies Poecilia mexicana (Poeciliidae, Teleostei) in light and darkness.
— Mem. Biospeol. 28: 163-167.
Plath, M., Parzefall, J. & Schlupp, I. (2003). The role of sexual harassment in cave- and
surface-dwelling populations of the Atlantic molly, Poecilia mexicana (Poeciliidae,
Teleostei). — Behav. Ecol. Sociobiol. 54: 303-309.
Plath, M., Brümmer, A. & Schlupp, I. (2004a). Sexual harassment in a livebearing fish
(Poecilia mexicana): influence of population-specific male mating behaviour. — Acta
Ethol. 7: 65-72.
96
Plath
Plath, M., Parzefall, J., Körner, K.E. & Schlupp, I. (2004b). Sexual selection in darkness?
Female mating preferences in surface- and cave-dwelling Atlantic mollies, Poecilia
mexicana (Poeciliidae, Teleostei). — Behav. Ecol. Sociobiol. 55: 596-601.
Plath, M., Arndt, M., Parzefall, J. & Schlupp, I. (2004c). Size-dependent male mating behaviour in the cave molly Poecilia mexicana (Poeciliidae, Teleostei). — Subterran. Biol. 2:
59-64.
Plath, M., Heubel, K.U., García de León, F.J. & Schlupp, I. (2005). Cave molly females (Poecilia mexicana, Poeciliidae, Teleostei) like well fed males. — Behav. Ecol. Sociobiol.
58: 144-151.
Plath, M., Brümmer, A., Parzefall, J. & Schlupp, I. (2006a). Size-dependent male mating
behaviour and sexual harassment in a population of Atlantic mollies (Poecilia mexicana)
from a sulphur creek. — Acta Ethol. 9: 15-21.
Plath, M., Seggel, U., Burmeister, H., Heubel, K.U. & Schlupp, I. (2006b). Choosy males
from the underground: Male mate choice in surface- and cave-dwelling Atlantic mollies,
Poecilia mexicana (Poeciliidae, Teleostei). — Naturwissenschaften 93: 103-109.
Plath, M., Makowicz, A.M., Schlupp, I. & Tobler, M. (2007a). Sexual harassment in livebearing fishes: comparing courting and non-courting species. — Behav. Ecol. 18: 680688.
Plath, M., Schlupp, I., Parzefall, J. & Riesch, R. (2007b). Female preference for large body
size in the cave molly, Poecilia mexicana (Poeciliidae, Teleostei): influence of speciesand sex-specific cues. — Behaviour 144: 1147-1160.
Plath, M., Blum, D., Schlupp, I. & Tiedemann, R. (2007c). Audience effect alters mating
preferences in Atlantic molly (Poecilia mexicana) males. — Anim. Behav. (in press).
Plath, M., Hauswaldt, J.S., Moll, K., Tobler, M., García de León, F.J., Schlupp, I. & Tiedemann, R. (2007d). Local adaptation and pronounced genetic differentiation in an extremophile fish, Poecilia mexicana, from a Mexican cave with toxic hydrogen sulfide.
— Mol. Ecol. 16: 967-976.
Plath, M., Tobler, M., Riesch, R., García de León, F.J., Giere, O. & Schlupp, I. (2007e).
Survival in an extreme habitat: the roles of behaviour and energy limitation. — Naturwissenschaften (in press).
Poulson, T.L. & Lavoie, K.H. (2000). The trophic basis of subterranean ecosystems. —
In: Ecosystems of the world: Subterranean ecosystems (Wilkens, H., Culver, D.C. &
Humphries, W.F., eds). Elsevier, Amsterdam, p. 231-249.
Proudlove, G.S. (2006). Subterranean fishes of the world. — International Society for Subterranean Biology, Moulis.
Reale, D., Bousses, P. & Chapuis, J.L. (1996). Female-biased mortality induced by male
sexual harassment in a feral sheep population. — Can. J. Zool. 74: 1812-1818.
Reznik, D.N. (1983). The structure of guppy life history: the trade-off between growth and
reproduction. — Ecology 64: 862-873.
Reznik, D.N. & Miles, D.B. (1989). A review of life history patterns in poeciliid fishes. —
In: Ecology and evolution of livebearing fishes (Poeciliidae) (Meffe, G.K. & Snelson,
F.F., eds). Prentice Hall, Englewood Cliffs, NJ, p. 125-148.
Ribowski, A. & Franck, D. (1993). Demonstration of strength and concealment of weakness
in escalating fights of male swordtails (Xiphophorus helleri). — Ethology 93: 265-274.
Riesch, R., Schlupp, I. & Plath, M. (2006a). Influence of male competition on male mating
behaviour in the cave molly (Poecilia mexicana, Poeciliidae, Teleostei). — J. Ethol. 24:
27-31.
Sexual harassment in extremophile mollies
97
Riesch, R., Schlupp, I., Tobler, M. & Plath, M. (2006b). Reduction of the association preference for conspecifics in surface- and cave-dwelling Atlantic mollies, Poecilia mexicana.
— Behav. Ecol. Sociobiol. 60: 794-802.
Rowe, L., Arnqvist, G., Sih, A. & Krupa, J.J. (1994). Sexual conflict and the evolutionary
ecology of mating patterns: Water striders as a model system. — Trends Ecol. Evol. 9:
289-293.
Ryan, M.J., Pease, C.M. & Morris, M.R. (1992). A genetic polymorphism in the swordtail
Xiphophorus nigrensis: Testing the prediction of equal fitnesses. — Am. Nat. 139: 2131.
Schlupp, I. & Plath, M. (2005). Male mate choice and sperm allocation in a sexual/asexual
mating complex (Poecilia mexicana, P. formosa, Poeciliidae, Teleostei). — Biol. Lett.
1: 169-171.
Schlupp, I., McKnab, R. & Ryan, M.J. (2001). Sexual harassment as a cost for molly females:
Bigger males cost less. — Behaviour 138: 277-286.
Shine, R., O’Connor, D. & Mason, R.T. (2000). Sexual conflict in the snake den. — Behav.
Ecol. Sociobiol. 48: 392-401.
Sibly, R.M. & Calow, P. (1989). A life-cycle theory of responses to stress. — Biol. J. Linn.
Soc. 37: 101-116.
Smith, C.C. (2007). Independent effects of male and female density on sexual harassment,
female fitness, and male competition for mates in the western mosquitofish Gambusia
affinis. — Behav. Ecol. Sociobiol. 61: 1349-1358.
Smith, C.C. & Sargent, R.C. (2006). Female fitness declines with increasing female density
but not male harassment in the western mosquitofish, Gambusia affinis. — Anim. Behav. 71: 401-407.
Smuts, B.B. & Smuts, R.W. (1993). Male aggression and sexual coercion of females in
nonhuman primates and other mammals: evidence and theoretical implications. — Adv.
Stud. Behav. 22: 1-63.
Stearns, S.C. (1989). Trade-offs in life-history evolution. — Funct. Ecol. 3: 259-268.
Sumner, I.T., Travis, J. & Johnson, C.D. (1994). Methods of female fertility advertisement
and variation among males in responsiveness in the sailfin molly (Poecilia latipinna).
— Copeia: 27-34.
Tobler, M., Schlupp, I., Heubel, K.U., Riesch, R., García de León, F.J., Giere, O. & Plath, M.
(2006). Life on the edge: Hydrogen sulfide and the fish communities of a Mexican cave
and surrounding waters. — Extremophiles 10: 577-585.
Tobler, M., Riesch, R., García de León, F.J., Schlupp, I. & Plath, M. (2007a). A new and
morphologically distinct population of cavernicolous Poecilia mexicana. — Environ.
Biol. Fish. (in press).
Tobler, M., Schlupp, I. & Plath, M. (2007b). Predation of a cavefish (Poecilia mexicana,
Poeciliidae) by a giant water bug (Belostoma, Belostomatidae) in a Mexican sulfur cave.
— Ecol. Entomol. 32: 492-495.
Tobler, M., Schlupp, I., García de León, F.J., Glaubrecht, M. & Plath, M. (2007c). Extreme
habitats as refuge from parasite infections? Evidence from an extremophile fish. — Acta
Oecol. 31: 270-275.
Valero, A., Hudson, R., Luna, E.A. & Garcia, C.M. (2005). A cost worth paying: energetically expensive interactions with males protect females from intrasexual aggression. —
Behav. Ecol. Sociobiol. 59: 262-269.
98
Plath
Walters, L.H. & Walters, V.W. (1965). Laboratory observations on a cavernicolous poeciliid
from Tabasco, Mexico. — Copeia: 214-233.
Weber, J.M. & Kramer, D.L. (1983). Effects of hypoxia and surface access on growth, mortality and behavior of juvenile guppies Poecilia reticulata. — Can. J. Fish. Aquat. Sci.
40: 1583-1588.
Zimmerer, E.J. & Kallman, K.D. (1989). Genetic basis for alternative reproductive tactics in
the pygmy swordtail, Xiphophorus nigrensis. — Evolution 43: 1298-1307.