Behavioral Ecology Advance Access originally published online on January 10, 2008
Behavioral Ecology 2008 19(3):463-474; doi:10.1093/beheco/arm139
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Female mating preferences for male morphological traits used in species and mate recognition in the Mexican sailfin mollies, Poecilia velifera and Poecilia petenensis
Department of Biological Sciences, 132 Long Hall, Clemson University, Clemson, SC 29634, USA
Address correspondence to M.B. Ptacek. E-mail: mptacek{at}clemson.edu.
Received 28 July 2006; revised 21 November 2007; accepted 21 November 2007.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONFUNDINGREFERENCES |
|---|
We tested whether sexually selected morphological traits in the sailfin mollies, Poecilia velifera and Poecilia petenensis, are also used in species recognition. Our first experiment, using live males as stimuli and providing females with olfactory as well as visual cues, found that females of both sailfin species preferred conspecific males to males of shortfin species. However, neither species preferred conspecific males when compared with heterospecific sailfin males, suggesting that premating reproductive isolation is not well developed between them. Our second experiment, providing females with only visual cues when distinguishing between live males, found that females of P. velifera preferred the larger of 2 stimulus males, regardless of whether the larger male was a conspecific or an heterospecific sailfin male. Such a preference for the larger sized male was not found in P. petenensis. To further investigate the role of the dorsal "sailfin" in species recognition, we used model males that varied only in the species identity of their dorsal fins. Females of both sailfin species preferred conspecific models with conspecific sailfins to those with dorsal fins of the shortfin species. In addition, females of P. velifera preferred the model with the largest sailfin, regardless of species identity. Similarly to the live male experiments, females of P. petenensis did not distinguish between conspecific and heterospecific sailfins. Overall, our study suggests that females of P. velifera have a generalized preference for larger males and that species-specific differences in sailfin shape do not lead to premating reproductive isolation between these 2 sailfin species.
Key words: female mating preferences, Poecilia petenensis, Poecilia velifera, sexual selection, species recognition.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONFUNDINGREFERENCES |
|---|
Mating signals can provide information to a receiver about species and individual identity, sex, readiness to mate, and mate quality (Andersson 1994
). Therefore, components of mating signals that are used in species recognition may also influence features that make the signaler more attractive to the receiver (Gerhardt 1982; Ryan and Wagner 1987; Ryan and Rand 1993; Boake et al. 1997; Ptacek 2000). Species recognition influences mate choice because matings with heterospecific males may result in low to no fitness gain because females may produce no offspring, low numbers of offspring, or offspring with reduced fitness (Gerhardt 1982; Harrison and Hall 1993; Noor 1995), and further, females may gain direct or indirect fitness benefits from choosing the highest quality mate (Andersson 1994; Pfennig 1998; Welch et al. 1998). Thus, to understand how female mating preferences can shape interspecific divergence in mating signals, it is necessary to ask if the same traits that are preferred by females within a species also function in species recognition (e.g., Gerhardt 1982; Boake et al. 1997; Pfennig 1998; Littlejohn 1999; Ptacek 2000).
A variety of male mating signals are multimodal, and disentangling the relative contributions of visual, chemical, and even auditory signals on species and mate recognition can be daunting. Studies in fishes, for example, have shown that signals from different sensory modalities can be conflicting. For example, in swordtail fishes (Xiphophorus spp.), female preferences for larger male size may have favored the evolution of elaboration of the caudal fin into a sword-like extension (Rosenthal and Evans 1998). Preference for swords has even led to females of a swordless species, Xiphophorus pygmaeus, preferring heterospecific males with swords to conspecific males that lack swords (Hankison and Morris 2002). However, when olfactory cues are available to these females, the preference for heterospecific males is abolished (Hankison and Morris 2003). Thus, studying how female mating preferences shape male mating signals must attempt to control for the multiple sensory modalities that females use when assessing male signals and the relevance of these sensory modalities to species and mate recognition.
The live-bearing fishes commonly known as sailfin mollies (Poeciliidae: Poecilia: Mollienesia) are ideal for studying the role of female mating preferences in shaping male mating signal divergence for several reasons. First, male sailfin mollies have a number of features of their mating signals that are known targets of female mating preferences. For example, all species of sailfin mollies perform a stereotyped courtship display that consists of a male approaching a female and raising and fanning the dorsal fin while bending the body in a C or sigmoid shape (Parzefall 1969, 1979; Farr 1989; Niemeitz et al. 2002; Ptacek et al. 2005). In 3 of the 4 sailfin species, the dorsal fin is greatly enlarged in males and is thought to enhance the visibility of the courtship display (Regan 1913; Hubbs 1933; Parzefall 1969) and females of at least one sailfin species prefer males with larger dorsal fins (MacLaren et al. 2004). Second, sailfin mollies form a monophyletic lineage diverging from shortfin molly ancestors (Ptacek and Breden 1998) that lack sexual dimorphism in dorsal fin size and do not perform courtship displays. Males of shortfin species instead rely on sneak copulations using gonopodial thrusts not preceded by courtship displays, and male–male competition for access to mates is a much stronger component of the shortfin mating system than is female choice (Farr 1989; Ptacek 1998). Thus, the enlarged dorsal fin, or sailfin, and courtship display behavior evolved in the common sailfin ancestor (Ptacek and Breden 1998; Ptacek et al. 2005; MacLaren and Rowland 2006a) and are thought to be important traits promoting the switch to a mating system where female mate choice is the primary factor determining mating success (Farr 1989; Ptacek 1998).
A third reason why female mating preferences may have been an important force shaping species divergence in mating signals in mollies relates to their geographic distribution patterns and phylogeographic history. Sailfin mollies are split into a freshwater lineage (Poecilia petenensis and Poecilia latipunctata) and a saltwater lineage (Poecilia velifera and Poecilia latipinna) (Ptacek and Breden 1998). Their current distributions are entirely allopatric with respect to one another, and even for the 2 species from the Yucatán peninsula of Mexico, P. velifera and P. petenensis, whose distributions are parapatric in the western portion of their ranges, the saltwater–freshwater habitat differences generally keep them from occurring sympatrically (Schmitter-Soto 1998; Hankison et al. 2006). Freshwater species of sailfin mollies (such as P. petenensis) do occur in sympatry with shortfin mollies, primarily Poecilia mexicana. The saltwater species, P. velifera, can co-occur with the shortfin Poecilia orri, where it has secondarily colonized freshwater cenotes, most likely as a result of flooding during recent hurricane events (Schmitter-Soto et al. 2002). Salt marsh populations of P. velifera, however, are not sympatric with shortfin molly species. Female mating preferences for the sailfin male phenotype appear to be ancestral, as females of shortfin species of mollies (P. mexicana and P. orri) prefer sailfin males to males of heterospecific shortfin species (Ptacek 1998) and models of conspecific males with enlarged dorsal fins (MacLaren and Rowland 2006a). Thus, a preexisting sensory bias (Basolo 1996; Endler and Basolo 1998) favoring larger sailfin males may have promoted initial divergence of sailfin mollies from shortfin ancestors.
Whether or not female mating preferences have been important in shaping mating signal divergence among the 4 sailfin molly species has not been investigated. Males of different sailfin species differ in the size and shape of their dorsal fins and in the rates of courtship displays and other mating behaviors (Hankison et al. 2006; Hankison and Ptacek 2007). For example, dorsal fin size varies between 2 of the Mexican sailfin species as a function of the number of dorsal fin rays (P. velifera, 15–21 fin rays; P. petenensis, 12–16 fin rays), as well as an overall difference in dorsal fin area (P. velifera, average area of 836 mm2; P. petenensis, average area of 710 mm2), resulting in males of P. velifera having an 18% larger dorsal fin than those of P. petenensis (Miller 1983; Hankison et al. 2006). Males of these 2 species differ in courtship display rates as well, with males of P. petenensis having higher display rates than males of P. velifera (Hankison and Ptacek 2007). Thus, potential differences in patterns of female mating preferences between different sailfin species may explain in part divergence in their mating signals.
In this study, we examined the role of female mating preferences in promoting male mating signal divergence in sailfin mollies using 3 types of dichotomous choice designs. First, we asked whether females preferred conspecific males to heterospecific sailfin and shortfin males using live males that were matched for size as stimuli. These tests were conducted in choice tanks where females had access to multiple species-specific components of the mating signal including courtship displays, morphology including dorsal fin shape, and potential olfactory differences between species. These trials assessed whether females showed generalized preferences for sailfin males, regardless of species, or if females always preferred conspecific males based on some species-specific components of the male mating signal. Because these 2 sailfin species differ in overall male size due primarily to differences in sailfin size, we then focused our tests on visual signals only and the role of larger male body size versus larger dorsal fin size in species and mate recognition. First, we investigated the role of preference for larger male size in species recognition by pairing conspecific and heterospecific sailfin males that differed in size. These trials assessed whether preference for species-specific mating signal components could override a generalized preference for larger males. Next, we investigated the role of the enlarged sailfin in mate recognition by constructing model males that held species-specific body shape constant, while varying dorsal fin shape. These trials assessed whether species-specific size and shape of the dorsal fin are important in species recognition or whether females showed a generalized preference for the model with the larger dorsal fin, regardless of species-specific shape differences. The importance of our study lies in its ability to tease apart various components of the mating signal and ask whether traits that are known targets of mate recognition (e.g., larger body size, larger dorsal fin size) also function in species recognition and premating behavioral isolation between different sailfin molly species.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONFUNDINGREFERENCES |
|---|
Study organisms
Fish used in this study were collected from wild populations in Mexico: P. velifera (Yucatán 1 population: 21°21.561‘N, 89°06.072’W) in June 2002, P. petenensis (Campeche 1 population: 19°08.620‘N, 90°57.400’W) in May 2001, and P. mexicana (Campeche 2 population: 19°14.230‘N, 90°50.110’W) in June 2002. The P. mexicana and P. petenensis populations used in this study were sympatric with respect to one another (although P. mexicana used in choice tests came from a different population than P. petenensis females), but P. velifera came from a salt marsh population that did not co-occur with any other species of molly.
Populations of all 3 species were maintained in separate Rubbermaid stock tanks (600 L; Fairlawn, OH) in 6 ppt saltwater, at a temperature of 25 °C and under ambient light conditions in a greenhouse in Clemson, SC. The stock populations were maintained at a density of approximately 150 breeding adults plus juveniles. Housing conditions followed those approved by the Clemson University Animal Research Committee (Protocol No. 50069).
Subjects used in the mate-choice trials were moved to Clemson University's Aquatic Animal Research Laboratory at least 2 weeks prior to testing. Fish were maintained in 3.75-L aquaria in 6 ppt saltwater, at a temperature of 25 °C and on a photoperiod of 14:10 light:dark, provided by Sylvania Gro-lux fluorescent bulbs (20-W, full spectrum 350–750 nm, with spectral peaks at 400, 440, and 540 nm; Denver, MA). Fish were fed once daily with commercial flake food (Ocean Star International Freshwater Flake (60%) with brine shrimp flake (38%) and spirulina flake (2%) mixture) (Burlingame, CA) (Clemson University Animal Research Protocol No. 40068). Female subjects used in the study were housed individually in aquaria, and stimulus males were housed in aquaria with a single female of the same species (female not used as a test subject). Adult males and females used in this study had been reared in captivity for more than 3 years, thus, were progeny from 2 to 3 generations of laboratory rearing.
Test females
We used dichotomous choice tests in 3 different experimental series to assess female mating preferences for particular live male or model male phenotypes. Female preference was measured as association time with each live male or model male presented in the choice arena during a 10-min observation period for each trial. A different group of females was used in each experiment: experiment 1, N = 24 for each species; experiment 2, N = 22 for P. velifera and N = 17 for P. petenensis; experiment 3, N = 24 for each species. Females used in all 3 experiments were controlled for receptivity by maintaining them individually in 3.75-L aquariums until they had released a brood and testing them within 24–48 h after partum. Females of the sailfin species P. latipinna have been found to be receptive to male attention only during this period of time (Farr and Travis 1986; Travis and Woodward 1989; Sumner et al. 1994), showing differential preferences in choice tanks, whereas gravid females do not (Ptacek and Travis 1997; Ptacek 1998; Gabor and Page 2003). Partitioned tanks are the only means in mollies to assess mating preferences without the confounding effects of male–male competition. In natural populations of the sailfin molly P. latipinna, behavioral observations have confirmed that females preferentially associate with larger males and males increase rates of all mating behaviors when approached by receptive females (Travis 1994).
Stimulus live or model males
Live males or models (cut out photographs) were presented to test females in each experiment as pairs matched for standard length (within ±5 mm SL). These stimulus males or models were kept within ±10 mm of the test female's SL. Stimuli between the 3 experiments differed in the overall male phenotype being assessed: experiment 1, species-specific traits of morphology (controlled for size differences), behavior, and olfactory signals (uncontrolled); experiment 2, morphology (controlled for size differences) and behavior signals (uncontrolled); experiment 3, morphology signals (controlled for dorsal fin size differences). Because we had a limited number of males in the appropriate size (SL) range for each experiment, some stimulus males or models had to be used more than once in different female-choice trials. In experiment 1, 7–16 different pairs of males (depending on male species combination) were used as stimulus males, with no more than 3 different test females viewing the same stimulus pair. In experiment 2, 13 different pairs of males were used as stimulus males across both male size combinations, again, with no more than 3 different test females viewing the same stimulus pair. In experiment 3, 8 different males of each species were photographed and used as conspecific models that varied in dorsal fin species characteristics. Thus, 3 females of the 24 tested for each sailfin species viewed the same pair of stimulus model males. In order to examine the potential for pseudoreplication due to the use of a stimulus set for more than 1 test female, we tested for a male stimulus pair effect on female preference using analysis of variance (ANOVA) (see statistical analyses below).
Experiment 1: female preference for conspecific males
This experiment used live male stimulus pairs presented in a dichotomous choice tank consisting of a 75-L aquarium (122 x 32 x 52 cm) with light above the tank provided by a single Sylvania Gro-lux fluorescent bulb (20-W, full spectrum 350–750 nm, with spectral peaks at 400, 440, and 540 nm). The tank was separated into 3 sections; 2 outer sections (24 cm) were the male compartments and were separated from the rest of the tank by clear Plexiglas containing slits at the top that allowed water to pass through to the central section. Thus, females could evaluate 3 different components of mating signals in these trials, differences in species of males based on morphology, courtship displays, and olfactory cues, the latter are known to be important in species recognition in other poeciliids (McLennan and Ryan 1997, 1999; Hankison and Morris 2003). The central section of the tank (74 cm) functioned as the female compartment. Lines on the front of the tank separated the central section into 3 zones: 2 preference zones, 1 on each side, that were each 10 cm from the male compartment, and a neutral zone (54 cm) in the middle. Such a narrow preference zone (ca. 1 body length) insured that when females were within this zone, they were looking at, and actively associating (following males along the partition or even attempting to position their gonopore in front of the male by backing into the partition) with stimulus males, rather than entering the zone by random chance. The male compartments were far enough apart (74 cm) that we did not observe the behavior of one male influencing the behavior of the other male during a trial, either when the 2 males were acclimating without the test female or in the presence of the test female. Males of both sailfin species (P. velifera and P. petenensis) performed courtship displays to both conspecific and heterospecific females (Kozak H, Cirino L, personal observation), but the number of displays performed by object males was not quantified during trials. It was observed however, in general, that many fewer displays were performed by these stimulus males in the partitioned choice tanks than are performed to females in direct contact trials (Hankison and Ptacek 2007). Thus, differences in courtship display rates between the 2 sailfin species and even in comparison to the shortfin P. mexicana males were minimized by the partition tank design. The outer sides and back of the choice tank were covered by black paper to reduce reflection inside the male compartments, and the front of the tank was covered with a 1-way film (Gila brand privacy window film, model PRS361, Martinsville, VA) to minimize disturbance to the fish by the presence of the observer. The tank was drained, rinsed with freshwater, and refilled after each trial to eliminate any olfactory cues left behind by previous object males.
To begin a trial, a stimulus male was placed in each of the end compartments of the choice tank and given a 15-min period of acclimation. The female was then placed in the center compartment and also allowed to acclimate for a 15-min period. This was followed by behavioral observations for a 10-min (600 s) period where the time in seconds that the female spent in each preference zone with each male and in the neutral zone, or central portion of the tank (apathy level), was recorded using stopwatches.
Each female was tested in 3 consecutive trials, each trial being 1 of 3 different male combinations. The male combinations were presented to females in a repeated-measures, Latin square design, which allowed us to test for any effects of order of trial and order in which the male combinations were presented. Males in each combination were matched for SL; thus, stimulus males differed in size only as a result of species-specific differences primarily in dorsal fin size and, to a lesser extent, caudal fin size and body depth (Figure 1). For each pair of stimulus males, we calculated the overall lateral projection area (LPA: the sum of body and dorsal and caudal fin areas) ratio between them (larger LPA/smaller LPA) from digital photographs of the left side of each anesthetized (0.1% MS-222) male with dorsal and caudal fins fully extended. To assess generalized preference for the sailfin male phenotype, we used 2 different male stimulus combinations: 1) a conspecific or 2) an heterospecific sailfin male paired with a male of a shortfin species, P. mexicana. A third combination assessed the strength of species recognition by pairing a conspecific male with an heterospecific sailfin male. We controlled for side bias by switching sides in which a particular species of male was presented between different male combinations.
|
Each female was tested in 3 consecutive trials; studies have shown that a female's responsiveness is not significantly affected across multiple trials per day (Ptacek and Travis 1997
; Ptacek 1998; MacLaren et al. 2004). Behavioral observations for these trials were made between 0800 and 1600 h from May through August 2004. Strength of preference (SOP) for each male combination was calculated by taking the difference in time spent with the 2 males; thus, we could compare SOP across different male combinations (Wagner 1998). The SOP was calculated by taking the time spent with the conspecific male minus the time spent with the heterospecific shortfin male (combination 1), the time spent with the conspecific male minus the time spent with the heterospecific sailfin male (combination 2), and the time spent with the heterospecific sailfin male minus the time spent with the heterospecific shortfin male (combination 3).
Experiment 2: female preference for larger males
Our second experiment used dichotomous choice tests with similar experimental procedures as those used in the first live male experiment to determine whether females had a preference for larger sailfin males regardless of their species identity. The choice tank was identical in size and configuration to that used in experiment 1, but in this experiment, the Plexiglas dividers were solid partitions. Thus, only visual signals could be used by females when comparing stimulus pairs of males. Because we were testing for the potential effects of larger male size overriding other species-specific visual signals, we wanted to eliminate any possible confounding effects of species differences in olfactory cues. Unfortunately, we could not control for differences in courtship displays between males used as stimulus pairs, either within or between species. However, as noted for experiment 1, overall rates of courtship displays in males of both sailfin species were low in the partitioned choice tank (Kozak H, Cirino L, personal observation).
Each female was tested with 2 different sailfin male combinations: 1) a large conspecific male paired with a small heterospecific sailfin male and 2) a small conspecific male paired with a large heterospecific sailfin male. The larger male used in the combination was 1.4–1.8 times larger than the small male used based on LPA. The SOP was calculated by taking the time a female spent with the larger male minus the time she spent with the smaller male. We alternated the order in which the combinations were presented between different females. The tank was drained, rinsed with freshwater, and refilled after each trial to eliminate any olfactory cues left behind by previous stimulus males. Behavioral observations for these trials were made between 0800 and 1600 h from October through April 2006.
Experiment 3: female preference for the sailfin
In this final experimental series, we wanted to test the influence of dorsal fin shape on female preference while holding all other aspects of the mating signal (courtship behavior, other morphological features, olfactory cues) constant. In order to manipulate a male's dorsal fin while holding all other aspects of conspecific male shape constant, we constructed model males (after design in MacLaren and Rowland 2006a; MacLaren et al. 2004) and used them in place of live males (Figure 1). Males of all 3 species (P. velifera, P. petenensis, and P. mexicana) that were used as stimulus males (body shape, dorsal fin shape, or both) were anesthetized (0.1% MS-222) and photographed using a Sony digital camera. We placed the anesthetized male on a dissecting tray and, using insect pins, pinned out the dorsal fin in a fully extended manner to simulate a courting male's LPA. Photographs were taken of the left side of each male. We then used the image analysis program Adobe Photoshop to vary dorsal fin shape to generate different sets of images from each original photograph. Dorsal fins of different species were placed on the photograph of the original male in the correct position (anterior to posterior axis) that they would be found on a similar sized male of each sailfin species and the shortfin species P. mexicana (Figure 1). We printed out the original and mirror images for each object model male onto transparency paper. An ink-jet printer was used to make the images, and color of the males appear as realistic as possible. However, transparency paper will not capture potential UV spectral reflection in these model males. We took the 2 sides of each image and attached them back-to-back with rubber cement glue. A piece of white paper was placed between the images (body area only) before attaching them together to keep the model male rigid (Rowland 1999; MacLaren et al. 2004).
These model males were then presented to receptive females in a dichotomous choice design to measure female preferences for species differences in sailfin size and shape. The test aquarium was divided into 3 sections by markings located on the front side of the tank, but the test female could freely move among all 3 sections. The 2 end sections were the preference zones (within 10 cm of each end of the aquarium), and the center section (64 cm) was considered the neutral or apathy zone. The model stimulus males were presented in the air outside of the female's choice tank, and the female viewed each model male through the end glass of her test aquarium. A thin wire was attached to the dorsal fin of each model, and the other end of the wire was attached to a belt positioned above and outside of the test aquarium. Each model male moved around its particular end of the aquarium by a motor and pulley system. After moving down the side of the tank, the model turned around 180° and "swam" back along the same side of the tank. This movement allowed the model to appear in a more realistic manner to that of a courting male than if the model had remained stationary, and this design has been shown to be effective in eliciting similar association times to that with live males in receptive females of the sailfin molly P. latipinna and the shortfin molly P. mexicana (MacLaren et al. 2004; MacLaren 2006; MacLaren and Rowland 2006a, 2006b).
To begin a trial, a female was placed in the testing tank and given 15 min for acclimation. We used opaque screens to block the view of the model males from the female during her acclimation period. We then removed the screens and turned on the pulley system so the model males began to move. After a 2-min acclimation period (for the female to view the moving stimulus pair of models), we began behavioral observations for a 10-min period. We recorded the time the female spent in each preference zone and in the center, neutral zone, using stopwatches. The same procedure was repeated for all the 3 male combinations with a 5-min break between trials. Behavioral observations for these trials were made between 0800 and 1600 h from January through August 2005.
Three male combinations were consecutively presented to each female. The male combinations were presented to females in a repeated-measures Latin square design, which allowed us to test for any effects of order of trial and order in which the male combinations were presented. Model males in each combination always had the conspecific body and caudal fin shape. Model males were equal in SL but varied in the species shape of the dorsal fin (Figure 1). To assess generalized preference for the sailfin, we used 2 different male combinations: 1) a model with a conspecific dorsal fin or 2) a model with an heterospecific sailfin dorsal fin paired with a model with a shortfin (P. mexicana) dorsal fin. A third combination assessed the strength of species recognition based on sailfin shape by pairing a model with a conspecific dorsal fin with a model with an heterospecific sailfin dorsal fin. Again, SOP was calculated for each of the 3 male combinations by taking the time spent with the model with the conspecific or heterospecific dorsal fin and subtracting the time spent with the model with the shortfin for the first 2 combinations and the time spent with the model with the conspecific dorsal fin and subtracting the time spent with the model with the heterospecific dorsal fin for the third combination.
Statistical analysis
For each of the 3 experiments, we first performed an ANOVA on SOP scores to examine the effects of presentation order (e.g., first, second, or third) and sequence order that the different male combinations were presented in (e.g., 1, 2, 3; 2, 3, 1; or 3, 1, 2). We then examined the residuals to verify that assumptions of parametric tests were met. We also used ANOVA to test for an effect of male (or model) stimulus pair on female SOP to show that different stimulus pairs of males elicited similar responses among test females. After verifying that there were no effects of order, sequence, or stimulus male pair and that residuals showed an even and broad distribution, we analyzed the SOP scores for experiments 1, 2, and 3 by performing a fully fit nested ANOVA, which included female species and male combination as fixed effects, and used female identity within a species as a block variable. We examined the effects of female species, male combination, female identity within species, and the interaction of male combination with female species; the female identity within species served as the residual error term for the main effect of female species. This analysis allowed us to test if there were differences between the 2 species in their overall patterns of female choice. Because male combination was significant in the first experiment, we conducted a post hoc comparison using Fisher's LSD to test for differences in SOP between the 3 male combinations.
Differences between the different male combinations in experiments 1 and 3 were due to both species-specific signal differences between object males and overall LPA differences. For example, differences between pairs of object males in LPA ratios were greatest between combinations containing sailfin and shortfin object males. Thus, we performed a second fully fit nested ANOVA with LPA ratio for each different pair of stimulus live or model males as a random effect and female species as a fixed effect and used female identity within a species as a block variable. We examined the effects of female species, LPA ratio, female identity within species, and the interaction of LPA ratio with female species; the female identity within species served as the residual error term for the main effect of female species.
To assess the level of female preference within each male combination for all 3 experiments, we tested for differences in the amount of time females spent with each male type for each of the different male combinations. These comparisons allowed us to discern whether females always spend more time with one particular type of male, regardless of which species or model type that male is paired with within choice tests. We performed these analyses on each species separately using paired t-tests and a sequential Bonferroni correction for multiple comparisons (Rice 1989).
To examine the effects of model males on overall levels of female preference and apathy (time in center), we performed an ANOVA on the combined data set of live and model male trials for both the SOP scores and the apathy levels. These data were partitioned into 7 sources: male combination, experiment type (live vs. model), female species, the interaction of experiment type and male combination, the interaction of female species and male combination, the interaction of female species and experiment type, and the 3-way interaction of female species, male combination, and experiment type. All statistical analyses were performed with SYSTAT Version 10 (Point Richmond, CA).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONFUNDINGREFERENCES |
|---|
Experiment 1: female preference for conspecific males
Females of each species spent more than half of the total trial time (600 s) interacting with object males (within 10 cm of either object male compartment) in all 3 male combinations. Apathy levels were fairly low, and females showed consistent preferences in some male combinations, although not every female tested showed the same direction of preference (Table 1).
|
Female SOP scores were not significantly affected by trial order (ANOVA: F1,135 = 0.156, P = 0.694) or the sequence in which the male combinations were presented (ANOVA: F4,135 = 1.353, P = 0.254). There was also no significant effect of male stimulus pair on female SOP scores for either species in any of the 3 male species combinations presented (P. velifera: combination 1, ANOVA: F1,10 = 2.686, P = 0.060; combination 2, ANOVA: F1,8 = 3.530, P = 0.057; combination 3, ANOVA: F1,14 = 1.152, P = 0.427; P. petenensis: combination 1, ANOVA: F1,10 = 0.848, P = 0.602; combination 2, ANOVA: F1,6 = 0.328, P = 0.910; combination 3, ANOVA: F1,15 = 1.922, P = 0.177). The ANOVA results showed that in 3 of the 4 sources, female species, female identity within a species, and the interaction between female species and male combination, there was no significant effect (Table 2). There was, however, a significant effect of male combination, indicating that SOP was greater in some male combinations than in others (Table 2). The Fisher's LSD post hoc comparison showed that female SOP in both species was greatest in the conspecific versus shortfin male combination (Fisher's LSD: P = 0.007).
|
The ANOVA results using the main effect of LPA ratio rather than male combination were not significant for any main effect or the interaction term (Table 2). Thus, females appeared to be using species characteristics of males to some degree when choosing particular males.
Although a significant male combination effect on female SOP suggests that females of the 2 species respond more to certain male species combinations than to others, it does not show the pattern of female preferences for particular male species across all 3 combinations. To examine this pattern in both species, we conducted paired t-tests to test for differences in association times of females with each species of stimulus male.
Females of both sailfin species showed preferences in some male combinations but the patterns were not consistent across the combinations within either species (Figure 2). For example, females of P. velifera had a significant preference (paired t-test: t = 4.37, P < 0.001, 
= 0.050) for conspecific males compared with shortfin males and there was a trend, though marginally nonsignificant after Bonferroni correction (paired t-test: t = 2.19, P = 0.039, = 0.025), for females to choose heterospecific sailfin males when compared with shortfin males (Figure 2A). Yet, there was no significant preference (paired t-test: t = 0.05, P = 0.962, = 0.017) for conspecific males when females compared them with heterospecific sailfin males.
|
Females of P. petenensis only showed a significant preference (paired t-test: t = 4.58, P < 0.001,
= 0.050) for conspecific males when compared with shortfin males. Although females of P. petenensis spent on average more time with heterospecific sailfin males when compared with the shortfin males (Figure 2B), this association was not a significant preference (paired t-test: t = 1.94, P = 0.065, = 0.025) and was much weaker than that shown by females of P. velifera in the equivalent male treatment combination (Figure 2A). Similarly to P. velifera, females of P. petenensis also did not have a significant preference (paired t-test: t = 0.88, P = 0.388, = 0.017) for conspecific males when compared with heterospecific sailfin males. Thus, neither species showed evidence of premating reproductive isolation across all species combinations based on female mating preferences.
Experiment 2: female preference for larger males
Female SOP scores were not significantly affected by the sequence in which the male combinations were presented (P. velifera, paired t-test: t = 0.757, P = 0.458 combination 1 first trial, t = 1.398, P = 0.178 combination 2 first trial; P. petenensis, paired t-test: t = 0.547, P = 0.592 combination 1 first trial, t = 0.585, P = 0.568 combination 2 first trial). There was also no significant effect of male stimulus pair on female SOP scores for either species (P. velifera: ANOVA: F1,12 = 0.793, P = 0.654; P. petenensis: ANOVA: F1,12 = 0.422, P = 0.937). The ANOVA results showed that in 3 of the 4 sources, male combination (ANOVA: F1,1 = 0.4457, P = 0.503), female identity within a species (ANOVA: F1,37 = 1.263, P = 0.241), and the interaction between female species and male combination (ANOVA: F1,1 = 0.234, P = 0.631), there was no significant effect. The 2 species of females did, however, differ in their SOP for larger males (ANOVA: F1,1 = 7.115, P = 0.020) with females of P. velifera having much greater SOP. Females of P. velifera showed consistent, significant (larger male conspecific, paired t-test: t = 4.45, P < 0.001, = 0.050; larger male heterospecific, paired t-test: t = 2.31, P = 0.016, = 0.025) preferences for the larger of the 2 object males, regardless of species (Figure 3A). The SOP for the larger male was not significantly different (paired t-test: t = 1.26, P = 0.222), regardless of whether the larger male was conspecific or heterospecific. Female preferences for the larger male in P. petenensis were overall much weaker and not significantly different (larger male conspecific, paired t-test: t = 1.60, P = 0.065, = 0.050; larger male heterospecific, paired t-test: t = 0.102, P = 0.460, = 0.025), regardless of which species of sailfin was the larger of the 2 stimulus males (Figure 3B). The SOP for the larger male was not significantly different (paired t-test: t = 1.57, P = 0.136), regardless of whether the larger males were conspecific or heterospecific. Thus, females of this species did not consistently prefer either the larger or the conspecific male.
|
Experiment 3: comparison of female responses to model males versus live males
Females of both species were responsive to model males and spent more than half of the total time (600 s) during trials interacting with them (within 10 cm of either end of the choice tank) in all 3 male combinations (Table 3). Apathy levels were low and similar to those seen during live male trials (Table 1). Females showed consistent preferences in some male combinations, but SOPs were lower in response to model male combinations than in live male combinations. Variation among females was higher in response to model males and not every female showed the same direction of preference for any of the 3 male combinations (Table 3).
|
The ANOVA for the comparison of SOP for combined live and model male experiments showed only 1 significant effect of the 7 sources, the interaction between experiment type and male combination (Table 4). This significant interaction was most likely due to the pattern of female preferences between the live and model male experiments in the male combination comparing the conspecific and heterospecific sailfin. Females, especially P. velifera, showed greater SOP for the larger, conspecific model male with the conspecific sailfin in the model male experiment (Table 3) than in the live male experiment (Table 1) for this male combination.
|
In the ANOVA comparing apathy levels for the 2 experiment types, there was, however, a significant effect of female species and the interaction between female species and experiment type (Table 4). The significant female species effect was due to the higher apathy levels found in P. petenensis females in both experiments. The interaction most likely resulted from the higher apathy levels of P. velifera females in response to model males, especially in the conspecific model with heterospecific sailfin versus conspecific model with shortfin combination (Table 3).
Experiment 3: female preference for the sailfin
The ANOVA for SOP using model males in the 3 male combinations showed no effect of trial order (ANOVA: F2,135 = 0.645, P = 0.527) or the sequence in which the male combinations were presented (F3,135 = 0.372, P = 0.774). There was also no significant effect of model male stimulus pair on female SOP scores for either species in any of the 3 model dorsal fin species combinations presented (P. velifera: combination 1, ANOVA: F1,7 = 1.976, P = 0.132; combination 2, ANOVA: F1,7 = 1.333, P = 0.298; combination 3, ANOVA: F1,7 = 0.665, P = 0.714; P. petenensis: combination 1, ANOVA: F1,7 = 1.036, P = 0.447; combination 2, ANOVA: F1,7 = 1.197, P = 0.363; combination 3, ANOVA: F1,7 = 1.268, P = 0.326). All 4 of the sources, male combination, female species, the interaction between female species and male combination, and female identity within a species, showed no significant effects on SOP (Table 5), thus females of both species were responding in a similar fashion to all 3 model male combinations. There was no significant main effect of LPA ratio on female SOP; however, the main effect of female species and the interaction of LPA ratio and female species were significant when performing the ANOVA using LPA ratio as a main effect rather than male combination (Table 5). Female SOP was weaker for P. petenensis in all model treatment combinations yielding the significant female species effect. The interaction between female species and LPA ratio was most likely due to the ability of females of P. velifera to "choose" the conspecific model with the "larger" sailfin in all 3 model male combinations, whereas females of P. petenensis only showed a preference for the conspecific model with the larger sailfin in 1 of the 3 combinations. Thus, female response to the conspecific model with the "larger" sailfin based on LPA ratio differences alone was much stronger for females of P. velifera than for females of P. petenensis.
|
The lack of a significant difference in SOP between the different male combinations could indicate that females of both sailfin species showed a generalized preference for a particular dorsal fin type (i.e., larger sailfin) regardless of species-specific shape differences. Alternatively, the lack of a male combination effect could indicate that females were not making choices between the different model male types. To distinguish between these 2 alternatives, we examined preferences within each male combination by comparing the association time that females spent with the different model males in each male combination using paired t-tests.
We found significant preferences when comparing time spent with each model male between combinations for females of both species (Figure 4). Females of P. velifera consistently preferred the model male with the larger dorsal fin (Figure 4A). This preference was significant in 2 of the 3 male combinations (conspecific models with conspecific vs. heterospecific sailfin, paired t-test: t = 2.77, P = 0.011, = 0.025; conspecific models with heterospecific sailfin vs. shortfin, paired t-test: t = 3.90, P = 0.001, = 0.050) and was marginally nonsignificant after Bonferroni correction (paired t-test: t = 2.49, P = 0.020, = 0.017) in the third combination between the conspecific models with conspecific sailfin and shortfin dorsal fins. Thus, females of this species appeared to base female preferences on sailfin size rather than species-specific shape components.
|
Unlike females of P. velifera that showed a strong trend toward always preferring the model with the largest dorsal fin, females of P. petenensis only showed a significant preference in 1 of the 3 model male combinations (Figure 4B). They showed a significant preference (paired t-test: t = 3.14, P = 0.005,
= 0.050) for conspecific models with conspecific sailfins when comparing them with conspecific models with shortfins. There was a weaker preference, though marginally nonsignificant after Bonferroni correction (paired t-test: t = 2.10, P = 0.047, = 0.025), for females to prefer conspecific models with the heterospecific sailfin to conspecific models with the shortfin dorsal fin. Unlike P. velifera females, P. petenensis females did not show a preference (paired t-test: t = 1.10, P = 0.284, = 0.017) for conspecific models with the dorsal fins of either sailfin species (conspecific or larger heterospecific sailfin) in the male combination that paired them. |
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONFUNDINGREFERENCES |
|---|
Two clear patterns emerge from the results of this study. First, species-specific differences in size and shape of the enlarged sailfin, the hallmark of the sailfin molly clade, do not serve as important species recognition signals for female sailfin mollies. In neither species of sailfin molly tested, did females prefer conspecific males to heterospecific sailfin males of equal size, nor did females prefer conspecific males when they were smaller than heterospecific sailfin males. Thus, the confounding influences of male size differences both within and between species conflict with potential species recognition signals involving the enlarged sailfin. The second conclusion drawn by this study is that females of P. velifera show much stronger preferences for larger males even when the larger male is an heterospecific sailfin species. Females of P. velifera consistently preferred larger conspecific males to shortfin males when matched for male size, the larger of 2 males regardless of sailfin species identity, and model males with the largest sailfin, again regardless of whether with conspecific or heterospecific sailfin shape. In contrast, females of P. petenensis only consistently preferred conspecific males or conspecific models with conspecific sailfins when comparing them with shortfin males or models. They showed no pattern of choice when comparing conspecific and heterospecific sailfins either matched for size or that varied in size. These results suggest that the 2 species differ with respect to the importance of larger male size as a mate-choice criterion.
Preference for larger males appears to be ancestral in sailfin mollies, as females of several shortfin species prefer sailfin males to heterospecific shortfin males (Ptacek 1998), and females of P. mexicana prefer conspecific model males with enlarged dorsal fins (MacLaren and Rowland 2006a). There are a number of potential reasons why female mollies, and poeciliids in general (e.g., Ryan and Keddy-Hector 1992; Ptacek and Travis 1997; Gabor and Page 2003), might prefer to associate with larger males. Studies have shown that because poeciliid males can sneak copulations through gonopodial thrusts, their constant attempts at forced copulations can lead to high levels of harassment for females resulting in decreased foraging efforts and potentially increased predation (Magurran and Seghers 1994; Schlupp et al. 2001). Larger males in the sailfin molly P. latipinna actively chase smaller males and prevent them from approaching large females (Travis 1994). Larger males are more likely to court females with displays than attempt constant gonopodial thrusts, at least in many populations of P. latipinna (Farr et al. 1986; Ptacek and Travis 1996). Thus, associating with larger males may afford some decrease in harassment from small males for females. Similarly, females may prefer to associate with larger fish in schooling aggregations. Whereas mollies only form loose schooling associations (Travis 1994; Ptacek M, personal observation), nonreceptive females have been shown to prefer to associate with both larger males and larger females in dichotomous choice tests (Gabor 1999).
Preference for larger males in mollies appears not to be the result of preference for specific traits, that is, larger male body size or larger dorsal fin size, but rather as a result of a preference for larger apparent male size through increased LPA (MacLaren et al. 2004; MacLaren 2006; MacLaren and Rowland 2006a). MacLaren et al. (2004) used model males that varied in body size, dorsal fin size, or both to show that larger models were more attractive to P. latipinna females whether due to increased body size or dorsal fin size. They concluded that the enlarged dorsal fin in male sailfin mollies may have evolved to achieve an increase in LPA when body size is constrained by other reasons, such as due to natural selection (MacLaren et al. 2004). A similar argument has been posed for the evolution of the sword in swordtails (Rosenthal and Evans 1998). Experimental manipulations using models have shown that larger males present a larger image on the female's retina, eliciting greater stimulation of the visual system and thus a stronger behavioral response (Rowland 1989a, 1989b; MacLaren et al. 2004; MacLaren 2006). This suggests that in sailfin mollies, preference for larger males is based on the size of the image projected on the female's retina, rather than the male's actual body size (MacLaren 2006).
So, if female preference for larger males is ancestral in mollies, and may be the result of a sensory bias for larger apparent male size, why then do females of P. petenensis seem to lack this strong preference for larger males? Whereas a number of studies have considered how preferences for elaborate male traits are acquired and maintained, far less attention has been paid to how female preferences, once acquired, might be lost or reversed (Wiens 2001). Reversal of preferences may, for example, occur if choice incurs a high cost (Kokko et al. 2003). Reversal of preference might also occur as the result of historical selection events operating against heterospecific matings (Pfennig 2000; Hankison and Morris 2002), as a response to environmental changes affecting signal perception (Seehausen et al. 1997), as a result of genetic drift (Rosenthal et al. 2002), and/or as a product of sexually antagonistic coevolution (Holland and Rice 1998). Although the underlying factors involved in the weakening of female preference for larger males in P. petenensis are unknown, one potential explanation may have to do with differences in the habitats occupied by these 2 sailfin species. The freshwater species, P. petenensis, occupies fast-flowing rivers in the interior of the Yucatán peninsula in Mexico, that have increased turbidity, especially during flooding events associated with the rainy season (Schmitter-Soto et al. 2002). Here constraints on sailfin size and shape for improved swimming performance and increased reliance on courtship displays to attract females may help to explain weaker preferences by females of this species for larger males. Indeed, morphological studies have shown that interpopulation variation in males of P. petenensis is greatest for characteristics of the caudal peduncle and caudal fin (both affect thrust and swimming speed; Webb 1978; Domenici 2003) whereas populations of P. velifera males vary most with respect to size and shape of the dorsal fin (Hankison et al. 2006). In addition, courtship display rates are higher in P. petenensis than P. velifera, and females might use courtship for more reliable mate recognition in turbid, fast-flowing habitats (Hankison and Ptacek 2007).
In addition to differences in habitat, P. petenensis is the species that likely evolved in sympatry with shortfin molly species, such as P. mexicana. Significant preferences for conspecific males when paired with shortfin males may have been due to discrimination against males of P. mexicana, a species that females of P. petenensis must regularly avoid heterospecific matings with. As a result of reinforcement of mating preferences in sympatry, females of P. petenensis may discriminate more strongly against "small" males; indeed males of P. petenensis less than 35 mm SL are almost nonexistent in natural populations (Hankison et al. 2006; Hankison and Ptacek 2007).
A final contribution of our study is with respect to the successful use of model males to test for differences in dorsal fin morphology on female mating preferences. Multimodal signals with a variety of components are difficult to dissect with respect to the influence of specific traits on female mating preferences. The advantage of using model males in mate-choice studies is that it allows for manipulation of 1 characteristic of a mating signal while holding all other components constant, a condition that is critical to assessing the importance of a mating signal component to either mate choice or mate recognition (Gerhardt 1982). In this study, females of both sailfin species showed similar SOP scores and apathy levels in response to model males as they did in trials using live stimulus males. Thus, model males offer a promising avenue for further studies to investigate the gain and loss of female mating preferences for particular components of molly mating signals.
The importance of female mating preferences in leading to premating reproductive isolation and speciation in sailfin mollies requires further exploration. Our study suggests that the enlarged dorsal fin characteristic of sailfin mollies, while potentially important in initial divergence of sailfins from shortfin ancestors, does not currently serve an important role in species recognition among sailfin species. Speciation of different sailfin species may instead be a result of allopatric divergence in different habitats (Miller 1983; Ptacek and Breden 1998), and changes in size and shape of the sailfin may be a by-product of adaptive divergence in morphology (Schluter 2001; Nosil et al. 2005), such as in response to swimming demands in the different environments in which the 2 species are found (salt marshes vs. rivers; e.g., Webb 1978; Walker 1984, 1997; Domenici 2003). Results of this study suggest that premating reproductive isolation between the allopatric sailfin species found in the Yucatán region of Mexico is not well developed, at least based on morphological differences between males of the 2 species in size and shape of the dorsal fin. Other mating signal components such as courtship behaviors, olfactory cues, and even male mating preferences (Gabor and Ryan 2001; Gumm and Gabor 2005; Gumm et al. 2006; Rafferty and Boughman 2006) require further study before the degree of premating reproductive isolation between these sailfin species can be fully assessed. Future studies should focus on the interactions between adaptive traits such as those used in swimming performance and traits that are targets of female mating preferences in promoting divergence in mating signals in Mexican sailfin mollies in order to assess the importance of both natural and sexual selection in their speciation.
|
FUNDING |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONFUNDINGREFERENCES |
|---|
Animal Behavior program (NSF grant IBN 0296173 to M.B.P.); Clemson University's Department of Biological Sciences (undergraduate research award to L.A.C.).
|
ACKNOWLEDGEMENTS |
|---|
We are especially grateful to Jason Thornton for building the model presentation apparatus. M.Childress S. Hankison, S. Gauthreaux, and 4 anonymous reviewers provided valuable comments on an earlier version of the manuscript. We thank Michele Kittell for fish care and M. Childress for advice and assistance with statistical analyses. We especially acknowledge the Mexican Government for granting permission to collect the fishes (Mexican Collecting Permit Nos 10.04.01.613.03 and 01.01.02.613.03 to M.B.P.).
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONFUNDINGREFERENCES |
|---|
Andersson M. Sexual selection. (1994) Princeton (NJ): Princeton University Press.
Basolo AL. The phylogenetic distribution of a female preference. Syst Biol (1996) 45:209–307.
Boake CRB, DeAngelis MP, Andreadis DK. Is sexual selection and species recognition a continuum? Mating behavior of the stalk-eyed fly Drosophila heteroneura. Proc Natl Acad Sci USA (1997) 94:12442–12445.
Domenici P. Habitat, body design, and the swimming performance of fish. In: Vertebrate biomechanics and evolution—Bels VL, Gasc JP, Casions A, eds. (2003) Oxford: Oxford BIOS Scientific Publishers, Ltd. 137–160.
Endler JA, Basolo AL. Sensory ecology, receiver biases, and sexual selection. Trends Ecol Evol (1998) 13:415–420.[CrossRef]
Farr JA. Sexual selection and secondary sexual differentiation in poeciliids: determinants of male mating success and evolution of female choice. In: Ecology and evolution of livebearing fishes—Meffe GK, Snelson FF Jr, eds. (1989) Englewood Cliffs (NJ): Prentice Hall. 91–123.
Farr JA, Travis J. Fertility advertisement by female sailfin mollies, Poecilia latipinna (Pisces:Poeciliidae). Copeia (1986) 1986:467–472.[CrossRef]
Farr JA, Travis J, Trexler JC. Behavioral allometry and interdemic variation in sexual behavior of the sailfin molly, Poecilia latipinna (Pisces:Poeciliidae). Anim Behav (1986) 34:597–609.
Gabor C. Association patterns of sailfin mollies (Poecilia latipinna): alternative hypotheses. Behav Ecol Sociobiol (1999) 46:333–340.[CrossRef][Web of Science]
Gabor C, Page R. Female preference for larger males in sailfin mollies, Poecilia latipinna: the importance of predation pressure and reproductive status. Acta Ethol (2003) 6:7–12.
Gabor CR, Ryan MJ. Geographical variation in reproductive character displacement in mate choice by male sailin mollies. Proc R Soc Lond B Biol Sci (2001) 268:1063–1070.[Medline]
Gerhardt HC. Sound pattern recognition in some North American treefrogs (Anura:Hylidae): implications for mate choice. Am Zool (1982) 22:581–595.[Web of Science]
Gumm JM, Gabor CR. Asexuals looking for sex: conflict between species and mate-quality recognition in sailfins mollies (Poecilia latipinna). Behav Ecol Sociobiol (2005) 58:558–565.[CrossRef][Web of Science]
Gumm JM, Gonzalez R, Aspbury AS, Gabor CR. Do I know you? Species recognition operates within and between the sexes in a unisexual-bisexual species complex. Ethology (2006) 112:448–457.[CrossRef][Web of Science]
Hankison SJ, Childress MJ, Schmitter-Soto JJ, Ptacek MB. Morphological divergence within and between populations of the Mexican sailfin mollies Poecilia velifera and P. petenensis. J Fish Biol (2006) 68:1610–1630.[CrossRef][Web of Science]
Hankison SJ, Morris MR. Sexual selection and species recognition in the pygmy swordtail, Xiphophorus pygmaeus: conflicting preferences. Behav Ecol Sociobiol (2002) 51:140–145.[CrossRef][Web of Science]
Hankison SJ, Morris MR. Avoiding compromise between sexual selection and species recognition: female swordtail fish assess multiple species-specific cues. Behav Ecol (2003) 14:282–287.
Hankison SJ, Ptacek MB. In review. Within and between species variation in male mating behaviors in the Mexican sailfin mollies Poecilia velifera and P. petenensis. Ethology.
Harrison JF, Hall HG. African-European honeybee hybrids have low nonintermediate metabolic capacities. Nature (1993) 363:258–260.[CrossRef]
Holland B, Rice WR. Perspective: chase-away sexual selection: antagonistic seduction versus resistance. Evolution (1998) 52:1–7.[CrossRef][Web of Science]
Hubbs C. Species and hybrids of Mollienisia. Aquarium (1933) 1:263–268. 277.
Kokko H, Brooks R, Jennions MD, Morley J. The evolution of mate choice and mating biases. Proc R Soc Lond B Biol Sci (2003) 270:653–664.[Medline]
Littlejohn MJ. Variation in advertisement calls of anurans across zonal interactions: the evolution and breakdown on homogamy. In: Geographic variation in behavior prespectives on evolutionary mechanisms—Foster SA, Endler J, eds. (1999) New York: Oxford University Press. 209–233.
MacLaren RD. The effects of male proximity, apparent size, and absolute size on female preference in the sailfin molly, Poecilia latipinna. Behaviour (2006) 143:1457–1472.[CrossRef]
MacLaren RD, Rowland WJ. Female preference for male lateral projection area in the shortfin molly, Poecilia mexicana: evidence for a pre-existing bias in sexual selection. Ethology (2006a) 112:678–690.[CrossRef][Web of Science]
MacLaren RD, Rowland WJ. Differences in female preference for male body size in Poecilia latipinna using simultaneous versus sequential stimulus presentation designs. Behaviour (2006b) 143:273–292.[CrossRef]
MacLaren RD, Rowland WJ, Morgan N. Female preference for sailfin and body size in the sailfin molly, Poecilia latipinna. Ethology (2004) 110:363–379.[CrossRef][Web of Science]
Magurran AE, Seghers BH. A cost of sexual harassment in the guppy. Proc R Soc Lond B Biol Sci (1994) 255:31–36.
McLennan DA, Ryan MJ. Reponses to conspecific and heterospecific olfactory cues in the swordtail Xiphophorus cortezi. Anim Behav (1997) 54:1077–1088.[CrossRef][Web of Science][Medline]
McLennan DA, Ryan MJ. Interspecific recognition and discrimination based upon olfactory cues in northern swordtails. Evolution (1999) 53:880–888.[CrossRef][Web of Science]
Miller RR. Checklist and key to the mollies of Mexico (Pisces: Poeciliidae: Poecilia, subgenus Mollienesia). Copeia (1983) 1983:817–822.[CrossRef]
Niemeitz A, Kreutzfeldt F, Schartl M, Parzefall J, Schlupp I. Male mating behavior of a molly, Poecilia latipunctata: a third host for the sperm-dependent Amazon molly, Poecilia formosa. Acta Ethol (2002) 5:45–49.[CrossRef]
Noor MA. Speciation driven by natural selection in Drosophilia. Nature (1995) 375:674–675.[CrossRef][Medline]
Nosil P, Vines TH, Funk DJ. Perspective: Reproductive isolation caused by natural selection against immigrants from divergent habitats. Evolution (2005) 59:705–719.[Web of Science][Medline]
Parzefall J. Zur vergleichenden Ethologie verschiedener Mollienesia-Arten einschliesslich einer Hohlenform von M. sphenops. Behaviour (1969) 33:1–37.[Medline]
Parzefall J. Zur genetick und biologischen bedeutung des aggressionsueshaltens von Poecilia sphenops (Pisces, Poeciliidae). Z Tierpsychol (1979) 50:399–422.[Web of Science]
Pfennig KS. The evolution of mate choice and the potential for conflict between species and mate-quality recognition. Proc R Soc Lond B Biol Sci (1998) 265:1743–1748.
Pfennig KS. Female spadefoot toads compromise on mate quality to ensure conspecific matings. Behav Ecol (2000) 11:220–227.
Ptacek MB. Interspecific mate choice in sailfin and shortfin species of mollies. Anim Behav (1998) 56:1145–1154.[CrossRef][Web of Science][Medline]
Ptacek MB. The role of mating preferences in shaping interspecific divergence in mating signals in vertebrates. Behav Process (2000) 51:111–134.[CrossRef][Web of Science][Medline]
Ptacek MB, Breden F. Phylogenetic relationships of the mollies (Poeciliidae: Poecilia: Mollienesia) based on mitochondrial DNA sequences. J Fish Biol (1998) 53:64–81.[Web of Science]
Ptacek MB, Childress MJ, Kittell MM. Characterizing the mating behaviors of the Tamesí molly, Poecilia latipunctata: a sailfin with shortfin morphology. Anim Behav (2005) 70:1339–1348.[CrossRef][Web of Science]
Ptacek MB, Travis J. Interpopulation variation in male mating behaviours in the sailfin mollie, Poecilia latipinna. Anim Behav (1996) 52:59–71.[CrossRef][Web of Science]
Ptacek MB, Travis J. Mate choice in the sailfin molly, Poecilia latipinna. Evolution (1997) 51:1217–1231.[CrossRef][Web of Science]
Rafferty NE, Boughman JW. Olfactory mate recognition in a sympatric species pair of three-spined sticklebacks. Behav Ecol (2006) 17:965–970.
Regan CT. A rebision of the cyprinodont fishes of the subfamily Poeciliinae. Proc Zool Soc Lond (1913) 11:977–1018.
Rice WR. Analyzing tables of statistical tests. Evolution (1989) 43:223–225.[CrossRef][Web of Science]
Rosenthal GG, Evans CS. Female preference for swords in Xiphophorus helleri reflects a bias for large apparent size. Proc Natl Acad Sci USA (1998) 95:4431–4436.
Rosenthal GG, Wagner WE Jr, Ryan MJ. Secondary reduction of preference for the sword ornament in the pygmy swordtail Xiphophorus nigrensis (Pisces: Poeciliidae). Anim Behav (2002) 63:37–45.[CrossRef][Web of Science]
Rowland WJ. Mate choice and the supernormality effect in female sticklebacks (Gasterosteus aculeatus). Behav Ecol Sociobiol (1989a) 24:433–438.[CrossRef][Web of Science]
Rowland WJ. The ethological basis of mate choice in male threespine sticklebacks, Gasterosteus aculeatus. Anim Behav (1989b) 38:112–120.[CrossRef][Web of Science]
Rowland WJ. Studying visual cues in fish behavior: a review of ethological techniques. Environ Biol Fishes (1999) 56:285–305.[CrossRef]
Ryan MJ, Keddy-Hector A. Directional patterns of female mate choice and the role of sensory biases. Am Nat (1992) 139:S4–S35.[CrossRef][Web of Science]
Ryan MJ, Rand AS. Species recognition and sexual selection as a unitary problem in animal communication. Evolution (1993) 47:647–657.[CrossRef][Web of Science]
Ryan MJ, Wagner WE Jr. Asymmetries in mating preferences between species: female swordtails prefer heterospecific males. Science (1987) 236:595–597.
Schlupp I, McKnab R, Ryan MJ. Sexual harassment as a cost for molly females: bigger males cost less. Behaviour (2001) 138:277–286.[CrossRef]
Schluter D. Ecology and the origin of species. Trends Ecol Evol (2001) 16:372–380.[CrossRef][Medline]
Schmitter-Soto JJ. Catálogo de los Peces Continentales de Quintana Roo. ECOSUR: San Cristóbal de Las Casas (1998).
Schmitter-Soto JJ, Comín FA, Escobar-Briones E, Herrera-Silveira J, Alococer J, Suárez-Morales E, Elías-Gutiérrez M, Díaz-Arce V, Marín LE, Steinich B. Hydrogeochemical and biological characteristics of cenotes in the Yucatán Peninsula (SE Mexico). Hydrobiologia (2002) 467:215–228.[CrossRef][Web of Science]
Seehausen O, van Alphen JJM, Witte F. Cichlid fish diversity threatened by eutrophication that curbs sexual selection. Science (1997) 277:1808–1811.
Sumner T, Travis J, Johnson CD. Methods of female fertility advertisement and variation among males in responsiveness in the sailfin molly (Poecilia latipinna). Copeia (1994) 1994:27–34.[CrossRef]
Travis J. Size-dependent behavioral variation and its genetic control within and among populations. In: Quantitative genetic approaches to animal behavior—Boake CRB, ed. (1994) Chicago (IL): University of Chicago Press. 165–187.
Travis J, Woodward BD. Social context and courtship flexibility in male sailfin mollies, Poecilia latipinna (Pisces:Poeciliidae). Anim Behav (1989) 38:1001–1011.[CrossRef][Web of Science]
Wagner WE Jr. Measuring female mating preferences. Anim Behav (1998) 55:1029–1042.[CrossRef][Web of Science][Medline]
Walker JA. Body form, locomotion and foraging in aquatic vertebrates. Am Zool (1984) 24:107–120.[Web of Science]
Walker JA. Ecological morphology of lacustrine threespinse stickleback Gasterosteus aculeatus (Gasterosteridae) body shape. Biol J Linn Soc (1997) 61:3–50.[CrossRef][Web of Science]
Webb PW. Fast-start performance and body form in seven species of teleost fish. J Exp Biol (1978) 74:211–226.
Welch A, Semlitsch R, Gerhardt HC. Call duration as an indicator of genetic quality in male gray tree frogs. Science (1998) 280:1928–1930.
Wiens JJ. Widespread loss of sexually selected traits: how the peacock lost its spots. Trends Ecol Evol (2001) 16:517–523.[CrossRef]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  J. L. Ward and D. A. McLennan Mate choice based on complex visual signals in the brook stickleback, Culaea inconstans Behav. Ecol., November 1, 2009; 20(6): 1323 - 1333. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||