Choosy males from the underground: male mating preferences in surface- and cave-dwelling Atlantic mollies (Poecilia mexicana)

Transcription

1 Naturwissenschaften (2006) 93: DOI /s z ORIGINAL ARTICLE Martin Plath. Uta Seggel. Heike Burmeister. Katja U. Heubel. Ingo Schlupp Choosy males from the underground: male mating preferences in surface- and cave-dwelling Atlantic mollies (Poecilia mexicana) Received: 24 August 2005 / Accepted: 10 November 2005 / Published online: 14 February 2006 # Springer-Verlag 2006 Abstract Atlantic mollies (Poecilia mexicana) inhabit a variety of surface habitats, but they also occur in a sulfur cave in southern Mexico. We examined male mate choice relative to female body size in the cave population and in the most closely related surface-dwelling population from a nearby river. Males from both populations were either lightor dark-reared and could choose between two differently sized females either on the basis of visual cues in light or on the basis of solely nonvisual cues in darkness. Sexual preferences were estimated from the degree of association. Cave molly males always showed a preference for the larger female, both in light and in darkness. Among the surface males, only light-reared males showed a preference in the visual cues test, but not in darkness. In a control experiment, we demonstrated that male association preferences directly translate into actual mating preferences. Apparently, using visual cues for mate choice is the ancestral state in this system, and using nonvisual cues has evolved as a novel trait in the cave population. We discuss the evolution M. Plath (*) Unit of Evolutionary Biology and Systematic Zoology, Institute for Biochemistry and Biology, University of Potsdam, Karl-Liebknecht Str , Potsdam, Germany martin_plath@web.de Tel.: Fax: U. Seggel. H. Burmeister Department of Behavior, Biozentrum Grindel, Martin-Luther-King Platz 3, Hamburg, Germany K. U. Heubel Department of Biological and Environmental Sciences, University of Helsinki, P.O. BOX 65 Helsinki 00014, Finland I. Schlupp Department of Zoology, University of Oklahoma, 730 Van Vleet Oval, Norman, OK 73019, USA of nonvisual male mate choice in the context of changed environmental conditions, namely the absence of light, hypoxia, and toxic hydrogen sulfide in the cave. Introduction Males may be choosy about the kind of females they mate with if females vary in reproductive quality (Andersson 1994; Berglund and Rosenquist 2001), such as the number of oocytes to be fertilized (Parker 1983). Males are predicted to show mating preferences if copulations are costly for males, for example, because available sperm is limited, because they need time to recover sperm stores after copulations (Nakatsuru and Kramer 1982; Verrell 1982), or because copulating is dangerous or energetically demanding. A prerequisite of male mate choice, however, is that males are able to perceive trait variation among females. In the present paper, we investigated male mate choice in the Atlantic molly (Poecilia mexicana), and we examined how the ability to perceive differences in female fecundity relates to male mating preferences in this species. To study this, we employed a comparative approach using two closely related P. mexicana populations from vastly different sensory environments. In live-bearing fishes (Poeciliidae), female fecundity is strongly correlated with body size (guppy, P. reticulata (Herdman et al. 2004; Ojanguren and Magurran 2004; Reznick and Endler 1982); mollies, P. latipinna (Travis et al. 1990); P. mexicana (Schlupp et al., unpublished data)). Therefore, males would benefit from mating with larger females, because they can father more offspring. Male mating preferences for large female body size have been demonstrated for guppies (Abrahams 1993; Herdman et al. 2004), mosquitofish (Gambusia holbrooki (Bisazza et al. 1989)), and sailfin mollies (P. latipinna (Gumm and Gabor 2005; Ptacek and Travis 1997)). One population studied originated from a typical river habitat ( surface form ), the other population ( cave molly ) inhabits an unusual molly habitat, the south Mexican Cueva del Azufre, a lightless sulfidic underground ecosystem

2 104 Fig. 1 a The sulfur creek outside the cave (El Azufre) harbors a huge population of Atlantic mollies. b The rearmost cave chamber (XIII) is separated from other cave chambers by a 1.5-m cascade. c A male from El Azufre with bright color patterns typical for surface-dwelling P. mexicana males. d Cave molly male caught in cave chamber XIII. Note that cave mollies still have eyes (Fig. 1). While many cave fishes are eyeless, cave mollies have retained slightly smaller but functional eyes (Peters et al. 1973). Studies on female mate choice have demonstrated specific adaptations to the changed environmental conditions in the cave molly. Females of the cave form express at least two mating preferences even in darkness (preference for large male body size (Plath et al. 2003a, 2004), preference for well-nourished males (Plath et al. 2005a)). Mating preferences based on nonvisual cues have not been found in light-reared surface-dwelling females from the nearby Rio Oxolotan, which showed a mating preference for large males only when visual cues from males could be assessed (Plath et al. 2003a, 2004) and showed no preference at all relative to male nutritional state (Plath et al. 2005a). A sensory shift seems to enable female mate choice in the cave form, whereby formerly visually mediated mating preferences are now mediated by nonvisual cues (e.g., water displacement, chemical or potentially acoustical cues). One aim of this study is to examine whether cave molly males show mating preferences in darkness in a fashion similar to cave molly females. Moreover, our previous studies on female preferences have raised the question on how plastic surface- and cave-dwelling P. mexicana are in using either visual or nonvisual cues for mate choice. We therefore asked whether different environmental conditions during ontogeny (exposure to or absence from light) influence which sensory channels are used by P. mexicana males to assess mate quality. To address this, both light- and dark-reared males from both populations were tested for a response to either visual or nonvisual cues of two differently sized females. Specifically, we asked whether P. mexicana males prefer larger females, and we examined whether the two populations studied differ in their response to visual or nonvisual cues. Especially interesting was whether cave molly males would show a preference for large females in darkness. In addition, we tested for an influence of light exposure during ontogeny on the use of visual and/or nonvisual cues for male mate choice in both populations. Methods Study organism and fish maintenance P. mexicana is widespread in Middle American freshwaters (Miller 1996). As a species with internal insemination, males use their modified anal fin, the gonopodium, to transfer sperm. P. mexicana males transfer only a part of the total sperm available per copulation (Schlupp and Plath 2005), and males may mate several times per day. The mating system is not resource-based. Broods are littered in an approximately monthly cycle; however, females can store sperm for several months. Thus, females require only one or a few copulations every few months (Constanz 1984). Nonetheless, males from surface habitats almost constantly attempt to mate, while females try to escape from this sexual harassment (Parzefall 1969; Plath et al. 2003b). A similar sexual conflict occurs in many poeciliids (e.g., guppies (Magurran 2001), sailfin mollies (Schlupp et al. 2001)). We caught mollies of two populations from adjacent but vastly different habitats in Tabasco, southern Mexico, near the village Tapijulapa (see Plath et al. (2005a)). The surface form inhabits a mostly clear, seasonally turbid swift river, the Río Oxolotan. The cave-dwelling population (cave molly) lives in the sulfidic Cueva del Azufre, which is situated on a plateau above the level of the river. The fish studied came from the innermost cave chamber (chamber XIII after (Gordon and Rosen 1962)), where the fish live constantly in

3 105 darkness. Based on a comparison of mitochondrial control region (D-loop) sequences of several P. mexicana populations, mollies from the Río Oxolotan appear to be the most recent ancestor of the cave form (Möller 2001). Hydrogen sulfide is toxic for fishes (Grieshaber and Völkel 1998). Moreover, H 2 S reacts with oxygen, which results in hypoxia. In the cave, H 2 S is used by sulfur bacteria for chemoautotrophic primary production. Sulfur bacteria, chironomids (Tendipes fulvipilus) and their larvae, as well as bat guano are the cave molly s primary food items (Langecker et al. 1996). Wild-caught cave molly males exhibit a significantly worse nutritional state than surfacedwelling ones (Plath et al. 2005a). Likely, sulfur bacteria represent abundant but low-energy food. To determine the relative influences of divergent evolution (evolution in a river habitat vs cave evolution) and ontogeny (exposure to light) on the use of visual or nonvisual cues during male mate choice, both populations were divided into two lineages: one half was reared under a 16:8-h light/dark illumination cycle, the other half was reared in a dark room. Only during feeding and fish care was dim illumination with visible light turned on for a few minutes in this room. Hence, even the dark-reared fish were to some extent familiar with light. This ensured that the dark-reared fish would behave calmly during the tests in light. The lineages were created in 1998 and were maintained in several mixed-sex stock tanks of l. Populations were not mixed. Test males (standard length, mean ± SE, 31.3±0.4 mm) were isolated from females in l tanks for 1 week before the trials. Meanwhile, illumination conditions were the same as those under which the males had been reared. Isolation from females for this time results in a drastic increase in the frequency of male sexual behaviors in P. mexicana (Franck 1975). Mate choice tests Males of each of the four lineages were tested for mating preferences with respect to female size in two independent treatments: The first tested for a response to solely visual cues; in the second treatment, only nonvisual cues were available. Each male was tested in only one experiment. Hence, eight separate experiments were conducted. Choice tests were conducted in a lightless room. The test tank ( cm) was visually divided in three compartments of equal size, two lateral preference zones and a neutral zone in the middle. Except for the front, all walls were covered by dark plastic foil. The bottom was covered with fine black gravel. A cylinder (12-cm diameter) was placed in both preference zones to hold the stimulus females. In the visual cues treatment, the cylinders were made of transparent Plexiglas, and the tank was illuminated by a 30-W neon tube 28 cm above the aquarium. In the nonvisual cues treatment, the cylinders were made of a plastic wire-mesh grid (5-mm mesh width, 1-mm wire diameter), which allowed chemical and water-pressurewave signals to pass through, and the tank was illuminated by a 500-W infrared bulb instead (λ>800 nm). In a previous study using microspectrophotometry, no photo pigments sensitive at more than 570 nm were found in the retina of P. mexicana (Körner et al., in press). Observations were conducted while the observer was sitting quietly 2.5 m from the aquarium (visual cues experiments) or with the help of an infrared-sensitive video camera (see Plath et al. (2004)), whereby the signal was transferred to a monitor in a neighboring room (nonvisual cues experiments). Before starting a trial, two females differing by size (mean ± SE, large 35.8±0.3 mm, small 26.2±0.3 mm, size difference 9.6±0.3 mm) were haphazardly taken from the stock tanks and were introduced in one of the two stimulus cylinders each (one on the left side and one on the right side). Sizes were within the range found in these populations (Plath et al., unpublished data). Females came from the same populations and lineages as the test males. All fish used in this study were sexually mature (females of these populations were found to develop eggs at approximately 23 mm length; Schlupp et al., unpublished data). After 10 min of habituation, a male was gently introduced in the neutral zone, and measurement of male preferences was initiated. We used association time spent near either kind of female (i.e., time inside the preference zones) as a measure of preference. After a first 10-min observation period, the cylinders were gently interchanged, and measurement was repeated likewise. This enabled us to detect side biases. All females were used only once per experiment; however, due to the limited number of fish in our stocks, stimulus females were sometimes used in both treatments (light and dark) but never in the same combination. Association times in a binary choice experiment are good predictors of actual mating preferences in guppy females (e.g., (Dugatkin and Godin 1992; Kodric-Brown 1993)), but it has been argued that association preferences need not be sexually motivated (Gabor 1999). We carried out a control experiment to test whether association preferences translate into actual mating preferences in this particular context. Light-reared cave molly males (34.2± 1.5 mm) were given an opportunity to mate with either a large (39.0±0.6 mm) or a small female (31.0±0.6 mm; size difference 8.1±0.6 mm). Some of the males used in this experiment had been used in the association preference tests before. However, experiments were 3 months apart; hence, we believe it extremely unlikely that experience from a previous test has influenced the results. Experimental evidence published by Witte and Massman (2003) indicated that females will not remember individual males after 48 h. The test tank was the same as during the association preference experiments in light. The observer was sitting quietly 1.5 m from the aquarium. Before a trial, a perforated Plexiglas cylinder was placed in the middle, and the test male was introduced into the cylinder. Two females of different size were introduced into the tank, and the three fish were given 10 min for habituation. Then, the cylinder was gently removed, and the following sexual behaviors of the male were recorded for 20 min: nipping, where the male touches the female genital opening to obtain gustatory information, and

4 106 Table 1 Male mate choice in a cave- and a surface-dwelling P. mexicana population Population Rearing Treatment Time near large female (s) Time near small female (s) Z N P A surface Light Visual 560 (249) 380 (195) ** Nonvisual 370 (145) 401 (177) Dark Visual 459 (211) 368 (214) Nonvisual 435 (78) 433 (69) B cave Light Visual 609 (123) 351 (165) <0.001*** Nonvisual 487 (126) 404 (106) * Dark Visual 568 (134) 402 (128) ** Nonvisual 422 (106) 381 (146) * Test fish were either light- or dark-reared and were given an opportunity to associate with a large or a small female under visible light (visual cues treatment) or under infrared conditions (nonvisual cues treatment). The time [median (interquartile range)] spent near either kind of female. Wilcoxon signed-rank tests, two-tailed. Significant differences are bold *P<0.05, **P<0.01, ***P<0.001 copulation attempts, where a male approaches a female laterally and attempts to introduce his gonopodium into the female genital opening (Parzefall 1969). The latter combined successful and attempted copulations, because in mollies, it is impossible to visually distinguish between actual copulations with or copulation attempts without sperm transfer (Schlupp and Plath 2005). Nipping typically, but not always, precedes copulations (Parzefall 1969). P. mexicana males show little or no courtship behavior (Parzefall 1969; Ptacek 2002), and in our experiments, no courtship displays have been observed. Statistical analyses To test for association preferences, times from the two parts of a trial were summed up. We decided a priori to exclude trials in which the male spent more than 80% of the choice time in only one preference zone (i.e., the male did not follow the initially preferred female) as side biases (17 out of 212 trials; with one to four excluded trials per treatment). Secondly, we decided to exclude trials if the male spent less than 50% of his total time in the preference zones. However, no trial had to be discarded due to low response. To test for male preferences within each experiment, the times spent near the larger and near the smaller female were compared using nonparametric Wilcoxon signed-rank tests. Furthermore, the relative time near the larger stimulus female was calculated as [time near large / (time near large + time near small )]. We used population as between factor and rearing conditions and testing conditions as within factors for a three-way analysis of variance (ANOVA) on arcsin-transformed data. Because the interaction term population rearing conditions testing conditions had no significant influence (F 1,187 =0.55, P=0.46), only interactions up to an interaction level of 2 were analyzed. To test for differences in the number of sexual behaviors in the full contact experiment, the frequency of sexual behaviors directed toward large or small females was compared using Wilcoxon signed-rank tests. Nipping and copulation attempts were analyzed separately. All P values are two-tailed. Fig. 2 The relative time near the large stimulus female in the different experiments. Riverdwelling males (a) and cave molly males (b) were given an opportunity to associate with a large or a small female. Males were either light-reared (left) or dark-reared (right) and were presented either visual cues (open bars) or nonvisual cues (gray bars) from the females. Box plots showing the median (middle line), the interquartile range (box), the 5 and 95% percentiles (whiskers). Open circles indicate data outside this range. Values >0.5 (dotted line) indicate preference for the larger female, values <0.5 indicate preference for the smaller female relative time near large female 0,8 0,7 0,6 0,5 0,4 A surface form B cave form visual cues non-visual cues light-reared dark-reared light-reared dark-reared

5 107 number of sexual behaviors Results Nipping Copulation attempts with large female with small female Fig. 3 The number of the sexual behaviors nipping (left) and copulation attempts (right) shown by cave molly males. A single male could interact with either a large (black bars) or a small female (white bars) in a simultaneous full contact choice experiment. For box plots, see Fig. 2 We successfully tested 195 males. In our statistical analysis, we first tested for a preference within each of the eight experiments. Cave molly males significantly preferred the larger female in all experiments (Table 1). Among surface males, only light-reared males significantly preferred the larger female in the visual cues treatment. Testing conditions had a significant influence on the fraction of time spent near the larger female (ANOVA: F 1,188 =25.39, P<0.0001). Overall, the males spent 59.68±1.42% of their time (mean ± SE) near the larger female during the tests in light but spent only 51.61±0.88% of their time near the large female in the absence of visible light (Fig. 2). The two populations differed in their response (F 1,188 =4.54, P=0.034); the males from the cave population showed a stronger preference (57.28±1.07% near the larger female) than surface males (53.76±1.36%). Rearing conditions had no statistically significant influence (F 1,188 =2.17, P=0.14); light-reared males spent 56.58±1.34% of their time near the larger female, and dark-reared males 54.41±1.09%. The interaction terms population rearing conditions (F 1,187 =0.74, P=0.39), population testing conditions (F 1,187 =0.19, P=0.67), and rearing conditions testing conditions also had no significant influence (F 1,187 =3.67, P=0.057). In the full contact experiment, cave molly males showed significantly more nipping with the larger of the two females (Wilcoxon signed-rank test: z= 2.98, P=0.0029, n=20; Fig. 3). Likewise, they also directed more copulation attempts toward the larger female (z= 2.22, P=0.026, n=20). Discussion The effect of the interaction term rearing conditions testing conditions was nonsignificant, suggesting that exposure to or absence of light during ontogeny only marginally influences the response to either visual or nonvisual cues: males that were exposed to light during ontogeny tended to show stronger preferences in the visual cues experiment. However, only cave molly males showed a preference in darkness, irrespective of rearing conditions. Given that the surface population examined is closely related with the cave form (Möller 2001), the visual preference has most likely been inherited from a river-dwelling common ancestor, but a novel way of assessing female size has evolved as a consequence of living in permanent darkness. In the full contact experiment, cave molly males directed more sexual behaviors toward larger females; hence, association preferences directly translate into mating preferences in this specific context. Mechanistically, chemical or mechanosensory cues may be used by cave molly males to assess female size differences in the absence of vision. Acoustical communication in poeciliids is unknown. Chemical communication by water-borne pheromones has been described for several poeciliids (e.g., (Hankinson and Morris 2003; McLennan and Ryan 1997, 1999; Shohet and Watt 2003)), but choice experiments using water containing odors of females vs control water failed to elicit a response in P. mexicana males (Wenzel et al., unpublished data). Potentially, the hydrogen sulfide in the cave water interferes with the use of waterborne pheromones in the cave, but this remains to be studied. The mechanosensory lateral line system has been shown to be involved in intersexual communication in other fishes (Satou et al. 1994). Compared with surface mollies, cave mollies have widened pores of the lateral line head canal system (Parzefall 2001). We propose that mechanical (water displacement) cues are being used to determine female size by using the lateral line. An unexpected outcome of this study was that cave molly males showed a slightly but significantly stronger preference than surface males (even in the visual cues treatment). Compared to surface-dwelling males, cave molly males show very little sexual behavior (mean ± SE, surface males 2.25±0.83 copulation attempts per minute, cave molly 0.17±0.12; (Plath et al. 2005b)), which is also true for laboratory-reared fish (Plath et al. 2003b). This is likely an adaptation to the extreme abiotic conditions like hypoxia, toxic H 2 S, and reduced food availability in their subterranean habitat. Generally, cave mollies show a genetically based reduction of energy-consuming behaviors like aggression to cope with these harsh conditions (Parzefall 2001). Cave mollies use aquatic surface respiration most of their time (Plath et al., unpublished data), but during mating, they cannot breathe at the water surface. It is tempting to speculate that cave molly males are choosier than surface males, because each copulation attempt is more energetically demanding for them.

6 108 Another possibility would be that selection has acted on cave molly females to be more sexually attractive to males, given that males are less sexually active. This hypothesis clearly warrants future investigations. Furthermore, males of surface-dwelling shortfin molly species like P. mexicana, P. sphenops, or P. gilii form dominance hierarchies, where some males attempt to exclude others from mating, which is believed to be an important determinant of male mating success (Farr 1989). In contrast, aggressive behavior and dominance hierarchies are absent in the cave molly (Parzefall 2001). Possibly, the potential for male mate choice is further increased in this system due to the absence of overt male competition. In summary, cave molly males use nonvisual cues to evaluate female quality (i.e., size), but surface-dwelling males do not. Using nonvisual cues for mate choice appears to be a novel trait in the cave population, which is not affected by different rearing conditions (light or darkness). Acknowledgements L. Fromhage and S. Nessler kindly read an earlier version of the manuscript. The fish team in Hamburg helped with animal care. The Mexican Government kindly issued permits to collect fish ( and DGOPA/5864/260704/-2408). Financial support came from the DFG to M.P. (Pl 470/1-1) and I.S. (SCHL 344/15-1). References Abrahams M (1993) The trade-off between foraging and courting in male guppies. Anim Behav 45: Andersson M (1994) Sexual selection. Princeton University Press, Princeton Berglund A, Rosenquist G (2001) Male pipefish prefer ornamented females. Anim Behav 61: Bisazza A, Marconato A, Marin G (1989) Male mate preference in the mosquitofish Gambusia holbrooki. Ethology 83: Constanz GD (1984) Sperm competition in poeciliid fishes. In: Smith RL (ed) Sperm competition and the evolution of animal mating systems. Academic, New York, pp Dugatkin LA, Godin J-GJ (1992) Reversal of female mate choice by copying in the guppy (Poecilia reticulata). Proc Biol Sci 249: Farr JA (1989) Sexual selection and secondary sexual differentiation in poeciliids: determinants of male mating success and the evolution of female choice. In: Meffe GK, Snelson FF Jr (eds) Ecology and evolution of livebearing fishes. Prentice Hall, Englewood Cliffs, pp Franck D (1975) Der Anteil des Coolidge-Effektes an der isolationsbedingten Zunahme sexueller Verhaltensweisen von Poecilia sphenops. Z Tierpsychol 38: Gabor C (1999) Association patterns of sailfin mollies (Poecilia latipinna): alternative hypotheses. Behav Ecol Sociobiol Gordon MS, Rosen DE (1962) A cavernicolous form of the poeciliid fish Poecilia sphenops from Tabasco, Mexico. Copeia 2: Grieshaber MK, Völkel S (1998) Animal adaptations for tolerance and exploitation of poisonous sulfide. Annu Rev Physiol 60:33 53 Gumm JM, Gabor CR (2005) Asexuals looking for sex: conflict between species and mate-quality recognition in sailfin mollies (Poecilia latipinna). Behav Ecol Sociobiol 58: DOI /s z Hankinson SJ, Morris MR (2003) Avoiding a compromise between sexual selection and species recognition: female swordtail fish assess multiple species-specific cues. Behav Ecol 14: Herdman EJE, Kelly CD, Godin J-GJ (2004) Male mate choice in the guppy (Poecilia reticulata): do males prefer larger females as mates? Ethology 110: Kodric-Brown A (1993) Female choice for multiple male criteria in guppies: interacting effects of dominance, coloration and courtship. Behav Ecol Sociobiol 32: Körner KE, Schlupp I, Plath M, Loew ER (in press) Spectral sensitivity of mollies: comparing surface- and cave-dwelling Atlantic mollies, Poecilia mexicana. J Fish Biol Langecker TG, 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: Magurran AE (2001) Sexual conflict and evolution in Trinidadian guppies. Genetica 112: McLennan DA, Ryan MJ (1997) Responses to conspecific and heterospecific olfactory cues in the swordtail Xiphophorus cortezi. Anim Behav 54: McLennan DA, Ryan MJ (1999) Interspecific recognition and discrimination based upon olfactory cues in northern swordtails. Evolution 53: Miller RR (1966) Geographic distribution of Central American freshwater fishes. Copeia 1966: Möller D (2001) Aspekte zur Populationsgenetik des eingeschlechtlichen Amazonenkärpflings Poecilia formosa (GIRARD 1859) unter Berücksichtigung der genetischen parentalen Art, dem Breitflossenkärpfling Poecilia latipinna (LESEUR 1821) und dem Atlantikkärpfling Poecilia mexicana, Steindachner Ph.D. thesis, University of Hamburg Nakatsuru K, Kramer DL (1982) Is sperm cheap? Limited malefertility and female choice in the lemon tetra (Pisces, Characidae). Science 216: Ojanguren AF, Magurran AE (2004) Uncoupling the links between male mating tactics and female attractiveness. Biol Lett 271: Parker GA (1983) Mate quality and mating decisions. In: Bateson P (ed) Mate choice. Cambridge University Press, Cambridge, pp Parzefall J (1969) Zur vergleichenden Ethologie verschiedener Mollienesiaarten einschließlich einer Höhlenform von Mollienesia sphenops. Behaviour 33:1 37 Parzefall J (2001) A review on morphological and behavioural changes in the cave molly Poecilia mexicana from Tabasco, Mexico. Environ Biol Fishes 50: Peters N, Peters G, Parzefall J, Wilkens H (1973) Über degenerative und konstruktive Merkmale bei einer phylogenetisch jungen Höhlenform von Poecilia sphenops (Pisces, Poeciliidae). Int Rev Gesamten Hydrobiol 58: Plath M, Körner KE, Parzefall J, Schlupp I (2003a) Persistence of a visually mediated mating preference in the cave molly, Poecilia mexicana (Poeciliidae, Teleostei). Subterr Biol 1:93 97 Plath M, Parzefall J, Schlupp I (2003b) The role of sexual harassment in cave- and surface-dwelling populations of the Atlantic molly, Poecilia mexicana (Poeciliidae, Teleostei). Behav Ecol Sociobiol 54: Plath M, Parzefall J, Körner KE, Schlupp I (2004) Sexual selection in darkness? Female mating preferences in surface- and cavedwelling Atlantic mollies, Poecilia mexicana (Poeciliidae, Teleostei). Behav Ecol Sociobiol 55: Plath M, Heubel KU, Garcia de Leon FJ, Schlupp I (2005a) Cave molly females (Poecilia mexicana, Poeciliidae, Teleostei) like well fed males. Behav Ecol Sociobiol 58: Plath M, Heubel KU, Schlupp I (2005b) Field observations on male mating behavior in surface- and cave-dwelling Atlantic mollies (Poecilia mexicana, Poeciliidae). Z Fischkd 7(2): Ptacek M (2002) Patterns of inheritance of mating signals in interspecific hybrids between sailfin and shortfin mollies (Poeciliidae: Poecilia: Mollienesia). Genetica 116:

7 109 Ptacek M, Travis J (1997) Mate choice in the sailfin molly. Evolution 51: Reznick D, Endler JA (1982) The impact of predation on life history evolution in Trinidadian guppies (Poecilia reticulata). Evolution 36: Satou M, Takeuchi HA, Nishii J, Tanabe M, Kitamura S, Okumoto N, Iwata M (1994) Behavioural and electrophysiological evidences that the lateral line is involved in the inter-sexual vibrational communicational of the hime salmon (landlocked red salmon, Oncorhynchos nerka). J Comp Physiol A 174: 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: Schlupp I, Mc Knab R, Ryan MJ (2001) Sexual harassment as a cost for molly females: bigger males cost less. Behaviour 138: Shohet AJ, Watt PJ (2003) Female association preferences based on olfactory cues in the guppy, Poecilia reticulata. Behav Ecol Sociobiol 55: Travis J, Trexler JC, Mulvey M (1990) Multiple paternity and its correlate in female Poecilia latipinna (Poeciliidae). Copeia 1990: Verrell PA (1982) Male newts prefer large females as mates. Anim Behav 30: Witte K, Massmann R (2003) Female sailfin mollies, Poecilia latipinna, remember males and copy the choice of others after 1 day. Anim Behav 65: