Key to predicting impacts of predation is understanding the mechanisms through which predators impact prey populations. While consumptive effects are well-known, non-consumptive predator effects (risk effects) are increasingly being recognized as important. Studies of risk effects, however, have focused largely on how trade-offs between food and safety affect fitness. Less documented, and appreciated, is the potential for predator presence to directly suppress prey reproduction and affect life-history characteristics. For the first time, we tested the effects of visual predator cues on reproduction of two prey species with different reproductive modes, lecithotrophy (i.e. embryonic development primarily fueled by yolk) and matrotrophy (i.e. energy for embryonic development directly supplied by the mother to the embryo through a vascular connection). Predation risk suppressed reproduction in the lecithotrophic prey (Gambusia holbrokii) but not the matrotroph (Heterandria formosa). Predator stress caused G. holbrooki to reduce clutch size by 43%, and to produce larger and heavier offspring compared to control females. H. formosa, however, did not show any such difference. In G. holbrooki we also found a significantly high percentage (14%) of stillbirths in predator-exposed treatments compared to controls (2%). To the best of our knowledge, this is the first direct empirical evidence of predation stress affecting stillbirths in prey. Our results suggest that matrotrophy, superfetation (clutch overlap), or both decrease the sensitivity of mothers to environmental fluctuation in resource (food) and stress (predation risk) levels compared to lecithotrophy. These mechanisms should be considered both when modeling consequences of perceived risk of predation on prey-predator population dynamics and when seeking to understand the evolution of reproductive modes.
Citation: Mukherjee S, Heithaus MR, Trexler JC, Ray-Mukherjee J, Vaudo J (2014) Perceived Risk of Predation Affects Reproductive Life-History Traits in Gambusia holbrooki, but Not in Heterandria formosa. PLoS ONE 9(2): e88832. https://doi.org/10.1371/journal.pone.0088832
Editor: Vincent Laudet, Ecole Normale Supérieure de Lyon, France
Received: April 12, 2013; Accepted: January 16, 2014; Published: February 13, 2014
Copyright: © 2014 Mukherjee et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: SM was supported by School of Environment, Arts and Society. Publication of this article was funded in part by Florida International University Open Access Publishing Fund. This material was developed in collaboration with the Florida Coastal Everglades Long-Term Ecological Research program under National Science Foundation Grant No. DBI-0620409. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Predators can affect prey populations through direct consumption (predation) as well as non-consumptive (“risk”) effects –. Risk effects often manifest through changes in foraging behavior and reduced access to energy that decreases reproduction . Less appreciated, but potentially important, is the possibility for physiological stress induced by predator presence to directly impact reproduction –.
Reproductive traits of prey individuals can be strongly influenced by predation, either by within-generation responses (plasticity, 9] or intergenerational change in genotypic frequencies presumably because of differential fitness (adaptive evolution, 10]. Examples of plasticity include field studies showing that prey tend to produce smaller clutches in habitats with predators or in the presence of a predation cue [e.g. killifish, Rivulus hartii, 11; song sparrow, Melospiza melodia, 12, 13]. Other plastic responses to predation risk include shortening of brood retention time [e.g. guppies, Poecilia reticulata, 14] and in some extreme cases completely foregoing reproduction [e.g. in Bank voles, Clethrionomys glareolus, 15]. Prey may show adaptive genetic differentiation in response to predator presence. In habitats with predators, prey may evolve to allocate more resources per individual offspring [e.g. guppies, Poecilia reticulata, 16]. These differences are maintained in common garden experiments, proving they are genetic rather than plastic.
Until now, however, there are relatively few experimental studies on predator-induced breeding suppression [PIBS, e.g. 15, 11, 17, 12, 13, see 18]. Physiological stressors such as predators are among the leading causes of life history variation, with prey responding to stress with higher concentrations of stress hormones (e.g. cortisol level), which in turn affects their immune systems and metabolism, causing suppressed digestion, growth and reproduction [see 3 and 8 for details]. Since PIBS likely affects predator-prey dynamics , , it is important to test its presence in a broader array of species and contexts.
When exposed to any environmental condition that makes breeding risky in terms of individual survival, behavioral and physiological responses should induce similar responses, namely a change in allocation of ‘effort’ away from reproduction and toward survival, at least in multi-brooded species . Behaviorally, this reallocation is manifested by reduced foraging effort in risky areas or at risky times [see examples in 1], thereby causing a net reduction in energy available to breed and thus to breeding effort. Physiologically, this reallocation makes stressed mothers invest relatively less overall energy in reproduction and relatively more in self-maintenance than would unstressed mothers , a standard result in birds and mammals. Thus, the physiological result would mimic energy limitation on reproductive allocation even when females are not proximately energy-limited. All else being equal, and if prey females (fishes in this study) are fed to satiation, we predicted that females exposed to predators should have smaller clutch sizes, and produce larger offspring, compared to those that were not exposed. This would suggest PIBS.
Reptiles and fishes display a diversity of modes of parental care and maternal investment that may affect how stress from predation risk cascades to affecting fitness. In principle, all prey species should be affected by predation risk (as explained above), however there maybe differences in responses due to differences in life history traits related to reproduction. For example, lecithotrophy and matrotrophy are two important reproductive modes. The lecithotrophs differ from matrotrophs is several ways. First, lecithotrophs invest all their reproductive energy in provisioning the eggs ‘up front’, while matrotrophs invest gradually . Secondly lecithotrophs have distinct reproductive episodes that have an ‘all or none’ quality to them, compared to the gradual reproductive output of matrotrophs. Thirdly, lecithotrophs provision their eggs over a much shorter timespan than do matrotrophs. Thus if predation risk were to favor reduced reproductive allocation in both species (see previous paragraph), one would expect to see more distinct tradeoffs between offspring size and number in lecithotrophs, simply because the total investment is limited and ‘paid’ out during a brief time period. Whereas in matrotrophs, investment can accumulate over time (while still being limited) and potential effects of stress (or energy allocation) on embryos can be adjusted through differential gestation length. We tested this hypothesis with two species of fish, Gambusia holbrooki (a lecithotroph) and Heterandria formosa (a matrotroph). Evaluating how prey species with different reproductive modes (i.e. with different patterns of embryo nourishment, e.g. matrotrophy versus lecithotrophy) may provide insight into how the costs of predation stress are borne by different species.
In this study we provide experimental evidence that predator-induced stress can differently affect reproduction in two species of poeciliid fishes with contrasting reproductive strategies (lecithotrophy and matrotrophy), suggesting that risk effects of predators may directly impact reproductive performance even in the absence of direct energetic costs of anti-predator behaviors, such as time allocation (where, when and how long to forage) and escape behavior. While the effect of predators on prey reproduction has been studied using several different predatory cues [e.g. visual and olfactory –15; auditory –13; chemical –22; simulated predator –23], we tested the effect of only a visual cue.
Research was carried out under IACUC Approval # 11-018 of Florida International University.
Experimental Fishes and their Reproductive Modes
Reproduction in vertebrates can be either oviparous (egg-laying) or viviparous (live-bearing). Viviparous gestation may be either lecithotrophy or matrotrophy, with both reproductive modes found in reptiles and fishes. In lecithotrophic organisms the embryonic development is fueled primarily from yolk, and the mother produces eggs that are stored internally until parturition . In contrast, a matrotrophic species produces ova, and supplements yolk energy after fertilization . Lecithotrophic females allocate energy and materials to developing embryos prior to fertilization, while matrotrophs are able to spread this investment throughout gestation . We used three species of freshwater fishes in our study. Gambusia holbrooki (eastern mosquitofish) and Heterandria formosa (least killifish), were the two prey fish species, that were visually exposed to their natural predator, the largemouth bass (Micropterus salmoides). G. holbrooki is a lecithotroph and lacks superfetation [clutch overlap; 24]. Though post-fertilization provisioning is known in Gambusia species [reviewed in 26], its contribution to embryo mass gain during development is small. For example, embryo mass in G. holbrooki decreases by as much as 20% or more [e.g., 27], while embryo mass in H. formosa increases by as much as 3000% , . Thus, G. holbrooki and H. formosa are representative of the continuum of lethotrophic and matrotrophic poeciliids , . Both fishes have relatively short lifespans in the Everglades (approximately a year) , .
G. holbrooki and H. formosa were collected from canals bordering the Everglades and ponds on Florida International University’s (FIU) Biscayne Bay Campus. The study did not involve endangered or protected species. Dip nets (5-mm mesh) were used to capture the fishes in the field which were transported to the lab in an insulted container fitted with an aerator. In the lab, fishes (stock) were held in single species groups (75.7 L tanks) for 30 days, before being housed individually for experiments. The fishes were fed ad libitum (see species specific food details below) in the morning and in the evening.
The experimental setup (Figure 1) consisted of six replicates of a central 75.7 L tank that contained either a single largemouth bass, Micropterus salmoides, (n = 3; standard length, SL: Mean ± SE = 171.1±2.6 mm) with a sponge filter (predator treatment), or a tank with only a sponge filter (control; n = 3). Six 18.9 L tanks (prey tanks, see Fig. 1) were positioned around the central tank, with each containing an adult female G. holbrooki (SL: Mean ± SD - Control: 27.55±3.01 mm; Predator: 27.20±2.23 mm) or H. formosa (SL: Mean ± SD - Control: 17.01±1.54 mm; Predator: 17.18±1.21 mm).
Each prey tank had an air-stone connected to an aerator, and was covered with white paper on three sides and along the bottom so that the prey fishes could not see movement in the room or other prey fish in adjacent tanks. An opaque white plastic sheet was also placed between the small tanks and the central tank (for both predator and control sets) so that the prey could not see the predator until the initiation of the experiment. All fish were kept on a 12∶12 h light:dark cycle, in a temperature controlled room maintained at 22°C.
Fishes were randomly selected from stock tanks and their weights and standard lengths measured before being housed individually in 18.9 L tanks. From the second day, the opaque partitions were removed twice a day, an hour each in the morning (9–10 AM) and late afternoon (4–5 PM), allowing the prey to see the predator or the control sponge filter. The air supply to the tanks was turned off during these hours. At the end of each hour, the partitions were returned and each prey tank was carefully searched for newborn offspring. As soon as an offspring was found, it was removed, euthanized using an ice slurry [33; approved by FIU’s IACUC], and its SL measured before being individually wrapped in aluminum foil and frozen. Later, offspring were dried for 3 days, and then weighed. Additionally, we also tracked the number of still born neonates present. Reproductive life-history data (litter size, offspring SL, offspring dry-mass) for the first 30 (for G. holbrooki) and 28 days (for H. formosa) from the start of the experiment were excluded from the analysis since these correspond to gestation time for each species [see 24, 34], and the young born during this time were likely conceived prior to the start of our experimental treatments.
After checking daily for newborns, the fish were fed. Each G. holbrooki received 0.05 ml of specially prepared liver paste , while each H. formosa received a chip (avg. weight - 4.1 mg) of TetraMin tropical crisps (Tetra Holding U.S. Inc.). Bass were fed with one pellet of Tetra Cichlid Jumbo Stick (Tetra Holding U.S. Inc.) at the end of each day. All prey fishes, regardless of treatment, were fed to satiation (indicated by remaining food) in the absence of predator cues. This allowed us to remove any potential effects of predators on energy intake, which might result in differences in reproduction. Experiments with G. holbrooki ran between May-August 2011, a time period that allowed us to collect two litters from most individuals, while experiments with H. formosa were conducted between September and October 2011.
Standard length of maternal fish was significantly correlated with litter size, but not with offspring SL or offspring dry weight in G. holbrookii. Hence we used a linear mixed model (random effects, with a repeated statement) in PROC MIXED , to analyze differences in litter size (length as a covariate), offspring SL and offspring dry weight, between treatment (control vs predator) and litter (first vs second litter). For H. formosa, we did not find a significant correlation between maternal SL and litter size, offspring SL or offspring dry weight. Since we had only two weeks of data for H. formosa, a one-way ANOVA was conducted to test for differences in the number of offspring, offspring SL and dry mass between treatment.
Differences in growth of the experimental (maternal) fishes were evaluated by measuring the SL of the fishes at the beginning and end of the experiment. With initial SL as a covariate, a one-way ANOVA was conducted to analyze differences in growth between treatments for G. holbrooki. For H. formosa, since there was no significant correlation between SL and growth, a one-way ANOVA was carried out. We used a t-test to calculate inter-brood interval (the number of days between 1st and 2nd litter) difference for G. holbrooki. All statistical analyses were conducted in SAS .
Exposure to predator visual cues modified reproductive parameters in G. holbrooki but not in H. formosa. G. holbrooki exposed to predators produced 43% fewer offspring compared to control individuals (Treatment: P = 0.02, Table 1; Effect size – Cohen’s d = 0.64, effect-size r = 0.31, Fig. 2a). Overall litter size of G. holbrooki remained the same between first and second litters (Litter: P = 0.64, Table 1). Interestingly, there was no effect of predator treatment in H. formosa. Control and predator exposed fishes produced similar number of offspring in both weeks of the experiment (P = 0.28, Table 1).
Red bar indicated predator treatment and blue bar indicate control. Alphabets in figures (b) and (c) indicate within litter differences.
Predator-exposed G. holbrokii produced offspring that were longer (Treatment; P = 0.01, Table 1; Effect size – Cohen’s d = 0.56, effect-size r = 0.27, Fig. 2b) and heavier (particularly in the first litter, Treatment*Litter; P = 0.06, Fig. 2c, Table 1). In the matrotrophic H. formosa, neither offspring standard length, nor dry mass was affected by predator treatments (Table 1).
Other Life-history Parameters
There was no statistical difference in growth between control and predator-exposed individuals in either G. holbrooki (ANCOVA, F1,32 = 0.05, P = 0.82) or H. formosa (ANOVA: F1,34 = 3.38, P = 0.08) during the experiment. Predation risk did not affect interbrood interval in G. holbrooki (t-test: t = 0.15, P = 0.89).
Our treatment affected the ratio of live vs stillbirth offspring for the lecithotrophic prey but not the matrotroph. While G. holbrooki mothers that were exposed to predators produced higher proportion (0.14) of stillborn compared to control mothers (0.02; 2×2 contingency Chi-square test, χ2 = 23.22, P<0.001; Fig. 3), no such difference was found in H. formosa (P = 0.75).
Predation risk may affect prey reproduction through two mechanisms. The first involves reductions because of foraging constraints when threatened by predators. In general, prey reduce their foraging rates, or alter habitat use, in order to enhance safety at the expense of energy intake. Such behavioral changes can result in less energy available for reproduction , . Alternatively, in species with parental care, individuals that must forage more cautiously when predators are present may reduce their visits to offspring, reducing offspring survival probabilities . Stress effects of predators may also be important to prey reproduction, even if prey are able to obtain similar amounts of energy compared to risk-free conditions because prey may alter their physiology to promote survival rather than reproduction , , , . We cannot exclude the possibility that our results resulted from differences in feeding, where the predator exposed fishes may have feed less, because we did not measure the amount of unconsumed food each day. However, since we did not find a difference in growth between predator and control fishes, and because predator treatment fishes had several predator (and competitor) free hours to feed, it is most likely that our observed reproductive impacts of predators arose through physiological (stress) responses. Future work should measure both unconsumed food and indicators of stress, such as cortisol.
Predation stress can affect reproduction in prey. Over two decades ago, Ylönen’s study  on bank voles (Clethrionomys glareolus) provided the first empirical evidence of breeding suppression by a predator through stress. Field data also suggest that stress effects of predation risk could be an important factor in population dynamics. For example, snow shoe hares (Lepus americanus) have high cortisol levels ,  and low testosterone responses in the presence of predators , while elk (Cervus elaphus) have lower progesterone levels that appear to reduce reproductive output in the presence of risk from wolves . Physiological effects of risk could also be maternally transmitted to offspring , , , which could further affect population dynamics.
Typically, seasonal breeders or species with short lifespan resist acute stress while favoring reproduction . Predator exposed G. holbrooki produced smaller litters, but larger offspring, likely trading off offspring quantity for enhanced mother survival in the face of stress. This is similar to the response of guppies (Poecilia reticulata) to food stress . Alternatively, the G. hobrooki mothers simply produced larger offspring that were more likely to survive when predators are around, i.e. optimizing offspring characteristics rather than degraded reproduction due to stress. The lack of response of H. formosa to predator treatment suggests resistance or insensitivity to stress. Reznick et al.  suggested that matrotrophic H. formosa were less able to modify offspring size in response to food level variation than lecithotrophic species such as P. reticulata. Since lecithotrophic species such as G. holbrooki must invest energy into offspring nourishment before fertilization, they may be more sensitive to stress effects than matrotrophs (i.e. after eggs have been yolked, 21, 43]. The fitness consequences of plasticity in response to stress are beyond the scope of this study, but are important to fully explore the implications of PIBS.
Differences in the response to predators can also be linked to how stress in transferred and regulated in the offspring. A positive relationship between maternal and egg glucocorticoid (stress hormone) concentrations have been found in fishes , . Though we did not measure stress hormone levels in mothers or offspring, we hypothesize that prey individuals exposed to predators produce higher levels of stress hormones, which altered metabolic costs of the developing embryos. Cortisol influences egg metabolic rate because it has a strong positive effect on metabolic rates in adult fishes . Early exposure to cortisol can also influence egg size and embryo survival , and also influences growth and development  and behavior . Future research should examine differences in the role of cortisol in regulating lecithotrophic and matrotrophic investment in prey offspring. Matrotropy may allow mothers to directly regulate (and reduce) stress levels in offspring. Hence, H. formosa may have transferred less stress to their offspring than the lecithotrophic G. holbrookii. Additionally, superfetation (i.e. clutch overlap), though not quantified in this study, may have also helped reduce the predatory stress effects on reproduction in the matrotrophic H. formosa. Thus, matrotrophy, superfetation, or both may allow mothers to better respond to environmental fluctuation in both resource (food) and stress (predation risk) levels than lecithotrophy.
There is a possibility that the difference in response between the two species is an artifact of the season when the experiments were conducted. G. holbrooki was tested during its prime breeding period, while H. formosa was tested later in the season at a time when natural populations of the species have either stopped breeding or are slowing considerably. Even though the conditions during the experiment were constant (day length and temperature), H. formosa individuals might still have been primed to expect a cessation of breeding. As noted above (see introduction), due to the life-history tradeoff between breeding effort and parental survival, at the end of a breeding season there is little future gain to withholding reproductive effort, even in the face of predation risk, because few females who are sexually mature at the start of winter survive to reproduce the following spring , , . Future studies should not only test H. formosa’s response during peak breeding season, but also G. holbrooki’s response later in the season.
Differences in reproductive strategy may also explain why G. holbrooki produced fewer and larger offspring under risk, while H. formosa did not. Since predation risk favors reduced reproductive allocation in both species, we expect stronger tradeoffs between offspring size and number in primarily lecithotrophic species such as G. holbrooki simply because the majority of investment is primarily made prior to fertilization at yolk, while in primarily matrotrophic species such as H. formosa, investment is spread over time and effects of stress on embryos can be adjusted throughout gestation. For G. holbrooki, which lacks superfetation and has relatively little post-fertilization maternally supplemented nourishment, this trade-off (offspring size vs number) may also be mediated by the ovarian space, and embryo packing determined at or soon after the yolking stage . Generally, predation risk declines with body size  and larger individuals are better in their escape abilities . For example, female sticklebacks (Gasterosteus aculeatus) have been shown to produce larger eggs (with high cortisol level) when exposed to predation , and this may be similar to what we found in G. holbrooki, if egg size is positively correlated with offspring size. Superfetation in H. formosa, on the other hand, may alleviate space constraints in the ovary , , leading to the differences in their life-history response to predation risk that we observed. Identifying the constraints on matrotrophy and evaluating it’s ‘adaptiveness’ remain a topic of debate .
Significantly higher proportion of stillbirths in predator exposed G. holbrookii compared to controls is consistent with a stress response, while the lack of differences in the matrotrophic H. formosa suggests a better ability to regulate stress in the developing offspring. Since the dead offspring were fully developed, these represent stillbirth (embryo death occurred in an advanced stage of development) rather than miscarriage (embryo death occurred in an early stage of development). These results are novel. Though previous experiments have shown that predation risk influences parental care (leading to greater proportion of failed hatchlings) and clutch size in prey organisms [e.g. 13], to the best of our knowledge no experimental study has shown predation risk to affect the number of stillbirths in a prey organism. This indicates yet another way that predation stress may affect population size of some prey species. The above results are also important in the context of the evolution of the placenta. According to the Trexler–DeAngelis  model, placentas should evolve in environments with consistently high levels of resource availability. An assumption highlighted by this model is that placental species abort embryos in low food conditions. This is the adaptive hypothesis, which assumes that the placenta evolves in response to some external ecological selection pressure in the environment , . A higher rate of stillbirths in predator-exposed G. holbrooki indicate that predation risk can also be an important external selective force (like food), with the potential to drive the evolution of the placenta in vertebrates.
We hypothesize that there is a relationship between mode of reproduction and response to predation risk, and future studies should test similar effects on several prey species, from each of the continuum of reproductive modes (oviparous, lecithotrohic and matrotrohic). Studies should not only test for the presence of PIBS in other predator-prey systems, but also conduct studies over a longer period to understand if PIBS is a short or long term life-history response. Ideally, these studies should also measure physiological and behavioral responses of prey [e.g.3, 4, 23] since they are vital for understanding the underlying mechanisms of these stress responses. Incorporating predation risk into life-history theory will not only provide important insights into life-history evolution , , but it will also help in understanding the implications of maternally-derived stress in an ecological context  and aid in elucidating the ecological impacts of changes in predator populations.
We thank - Elizabeth Huselid and Amanda Banet for advice on fish husbandry and poeciliid reproductive biology, and Allstate Resource Management Inc., Davie, for donating largemouth bass. SM would like to thank the School of Environment, Arts and Society for post-doctoral support. Authors would also like to thank four anonymous reviewers for their valuable comments which helped greatly improve this manuscript.
Conceived and designed the experiments: SM MH JT. Performed the experiments: SM JRM JV. Analyzed the data: SM JRM. Contributed reagents/materials/analysis tools: JT MH. Wrote the paper: SM. Critical comments on the manuscript: MH JT JRM.
- 1. Lima SL, Dill LM (1990) Behavioral decisions made under the risk of predation: a review and prospectus. Can J Zool 68: 619–640.
- 2. Pekarsky BL, Cowan CA, Penton MA, Anderson C (1993) Sublethal consequences of stream-dwelling predatory stoneflies on mayfly growth and fecundity. Ecology 74: 1836–1846.
- 3. Boonstra R, Hik D, Singleton GR, Tinnikov A (1998) The impact of predator-induced stress on the snowshoe hare cycle. Ecol Monograph 79: 371–394.
- 4. Creel S, Chritianson D, Liley S, Winnie Jr JA (2007) Predation risk affects reproductive physiology and demography of elk. Science 315: 960.
- 5. Creel S, Chritianson D (2008) Relationships between direct predation and risk effects. Trends Ecol Evolut 23: 194–201.
- 6. Clinchy M, Zanette L, Boonstra R, Wingfield JC, Smith JNM (2004) Balancing food and predator pressure induces chronic stress in songbirds. Proc R Soc Lond [Biol] 271: 2473–2479.
- 7. Sheriff MJ, Krebs CJ, Boonstra R (2009) The sensitive hare: sublethal effects of predator stress on reproduction in snow shoe hares. J Anim Ecol 78: 1249–1258.
- 8. Hawlena D, Schmitz OJ (2010) Physiological stress as a fundamental mechanism linking predation to ecosystem functioning. Am Nat 176: 537–556.
- 9. Magnhagen C (1991) Predation risk as a cost of reproduction. Trends Ecol. Evol. 6: 183–186.
- 10. Reznick DN, Ghalambor CK (2001) The population ecology of contemporary adaptations: what empirical studies reveal about the conditions that promote adaptive evolution. Genetica 112: 183–198.
- 11. Fraser DF, Gilliam JF (1992) Nonlethal impacts of predator invasion: facultative suppression of growth and reproduction. Ecology 72: 959–970.
- 12. Travers M, Clinchy M, Zanette L, Boonstra R, Williams TD (2010) Indirect predator effects on clutch size and the cost of egg production. Ecol Lett 13: 980–988.
- 13. Zanette LY, White AF, Allen MC, Clinchy M (2011) Perceived predation risk reduces the number of offspring songbirds produce per year. Science 334: 1398–1401.
- 14. Evans JP, Gasparini C, Pilastro A (2007) Female guppies shorten brood retention in response to predator cues. Behav Ecol Socio Biol 61: 719–727.
- 15. Ylönen H (1989) Weasels Mustela nivalis suppress reproduction in cyclic bank voles Clethrionomys glareolus. Oikos 55: 138–140.
- 16. Reznick DN, Endler JA (1982) The impact of predation on life history evolution of Trinidadian guppies. Evolution 36: 160–177.
- 17. Eggers S, Griesser M, Nystrand M, Ekman J (2006) Predation risk induces changes in nest-site selection and clutch size in the Siberian jay. Proc R Soc Lond [Biol] 273: 701–706.
- 18. Norrdahl K, Korpimäki E (2000) The impact of predation risk from small mustelids on prey populations. Mammal Rev 3& 4: 147–156.
- 19. Ruxton G, Lima SL (1997) Predator-induced breeding suppression and its consequences for predator-prey population dynamics. Proc R Soc Lond [Biol] 264: 409–415.
- 20. Sterns SC (1992) The evolution of life-histories. Oxford Univ. Press, New York.
- 21. Trexler JC (1997) Resource availability and offspring provisioning: plasticity in embryo nourishment in sailfin mollies. Ecology 78: 1370–1381.
- 22. Crowl TA, Covich AP (1990) Predator-induced life-history shifts in a freshwater snail. Science 247: 949–951.
- 23. Geising ER, Suski CD, Warner RE, Bell AM (2011) Female sticklebacks transfer information via eggs: effects of maternal experience with predators on offspring. Proc R Soc Lond [Biol] 278: 1753–1759.
- 24. Pyke GH (2005) A review of the biology of Gambusia affinis and G. holbrooki. Rev Fish Biol Fisheries 15: 339–365.
- 25. Turner CL (1940) Pseudoamnion, pseudochorion, and follicular pseudoplacenta in poeciliid fishes. J Morph 67: 59–89.
- 26. Marsh-Matthews E, Deaton R, Brooks M (2010) Survey of matrotrophy in lecithotrophic poeciliids. Pages 255–258 in: Viviparous Fishes II, the proceedings of the III International Symposium on Viviparous Fishes. (M. C. Uribe and H. J. Grier, editors). New Life Publications, Homestead, FL.
- 27. Edwards TM, Miller HD, Guillette Jr LJ (2006) Water quality influences reproduction in female mosquitofish (Gambusia holbrooki) from eight Florida Springs. Environ Health Perspect 114 (suppl 1)69–75.
- 28. Scrimshaw N (1945) Embryonic development in Poeciliid fishes. Biol. Bull. 88: 233–246.
- 29. Reznick DN, Cakkahan H, Llauredo R (1996) Maternal effects on offspring quality in Poeciliid fishes. American Zoologist 36: 147–156.
- 30. Pires MN, Banet AI, Pollux BJA, Reznick DN (2011) Variation and evolution of reproductive strategies. In J. P. Evans, A. Pilastro, and I. Schlupp, eds. Ecology and Evolution of Poeciliid Fishes. Univ Chicago Press. Pp 28–37.
- 31. Haake PW, Dean JM (1983) Age and growth of four Everglades fishes using otolith techniques. Technical Report SFRC83/03. South Florida Research Center, Everglades National Park. Homestead, FL. 68pp.
- 32. Konnert TJ (2002) The effects of hydroperiod on the life history parameters of Poecilia latipinna and Heterandria formosa (Poeciliidae) in the Florida Everglades. M.S. Thesis, Florida International University, Miami, 91pp.
- 33. Blessing JJ, Marshall JC, Balcombe SR (2010) Humane killing of fish for scientific research: a comparison of two methods. J Fish Biol 76: 2571–2577.
- 34. Leips JJ, Richardson ML, Rodd FH, Travis J (2009) Adaptive maternal adjustments of offspring size in response to conspecific density in two populations of the least killifish, Heterandria formosa. Evolution 63: 1341–1347.
- 35. Banet AI, Reznick DN (2008) Do placental species abort offspring? Testing an assumption of the Trexler-DeAngelis model. Funct Ecol 22: 323–331.
- 36. SAS Institute, 2008. SAS/STAT 9.2 user’s guide, SAS Institute Inc. Cary, North Carolina.
- 37. Creel S, Chritianson D, Winnie Jr JA (2009) Glucocorticoid stress hormones and the effect of predation risk on elk reproduction. Proc Natl Acad Sci USA 106: 12388–12393.
- 38. Wingfield J, Maney DL, Breuner GW, Jacobs JD, Lynn S, Ramenofsky M, Richardson RD (1998) Ecological bases of hormone-behavior interactions: the “emergency life history stage.”. Amer Zool 38: 191–206.
- 39. Clinchy M, Sheriff MJ, Zanette LY (2013) Predator-induced stress and the ecology of fear. Funct Ecol 27: 56–65.
- 40. Wingfield JC, Sapolsky RM (2003) Reproduction and resistance to stress: when and how. J Neuroendochrinol 15: 711–724.
- 41. Sheriff MJ, Krebs CJ, Boonstra R (2010) The ghosts of predators past: population cycles and the role of maternal programming under fluctuating predation risk. Ecology 91: 2983–2994.
- 42. Reznick DN, Yang AP (1993) The influence of fluctuating resources on Life history: patterns of allocation and plasticity in female guppies’. Ecology 74: 2011–2019.
- 43. Trexler JC, DeAngelis DL (2003) Resource Allocation in Offspring Provisioning: An evaluation of the conditions favoring the evolution of matrotrophy. Am Nat 162: 574–585.
- 44. McCormick MI (1998) Behaviourally induced maternal stress in a fish influences progeny quality by a hormonal mechanism. Ecology 79: 1873–1883.
- 45. Schreck CB, Contreras-Sanchez W, Fitzpatrick MS (2001) Effects of stress on fish reproduction, gamete quality, and progeny. Aquaculture 197: 3–24.
- 46. Wendelaar Bonga SE (1997) The stress response in fish. Physiol Rev 77: 591–625.
- 47. Eriksen MS, Bakken M, Espmark A, Braastad BO, Salte R (2006) Pre-spawning stress in farmed Atlantic salmon, Salmo salar: maternal cortisol exposure and hyperthermia during embryonic development affect offspring survival, growth and incidence of malformations. J Fish Biol 69: 114–129.
- 48. Espmark AM, Eriksen MS, Salte R, Braastad BO, Bakken M (2008) A note on pre-spawning maternal cortisol exposure in farmed Atlantic salmon and its impact on the behaviour of offspring in response to a novel environment. Appl Anim Behav Sci 110: 404–409.
- 49. Trexler JC, Travis J, McManus M (1992) Effects of habitat and body size on mortality rates of Poecilia latipinna. Ecology 73: 2224–2236.
- 50. Werner E (1986) Amphibian metamorphosis: growth rate, predation risk, and the optimal size at transformation. Am Nat 128: 319–341.
- 51. Schürch R, Taborsky B (2005) The functional significance of buccal feeding in the mouth brooding cichlid Tropheus moorii. Behaviour 142: 265–281.
- 52. Travis J, Farr JA, Henrick S, Cheong RT (1987) Testing theories of clutch overlap with the reproductive ecology of Heterandria formosa. Ecology 68: 611–623.
- 53. Zúñiga-Vega JJ, Reznick DN, Johnson JB (2007) Habitat predicts reproductive superfetation and body shape in the livebearing fish Poeciliopsis turrubarensis. Oikos 116: 995–1005.
- 54. Pollux BJA, Reznick DN (2011) Matrotrophy limits a female’s ability to adaptively adjust offspring size and fecundity in fluctuating environments. Functional Ecology 25: 747–756.
- 55. Thibault R, Schultz RJ (1978) Reproductive adaptations among viviparous fishes (Cyprinodontifomes: Poeciliidae). Evolution 32: 320–333.
- 56. Trexler JC, DeAngelis DL, Jiang J (2011) Community assembly and mode of reproduction: predicting the distribution of livebearing fishes. In J. P. Evans, A. Pilastro, and I. Schlupp, eds. Ecology and evolution of poeciliid fishes. University of Chicago Press, Chicago. Pp 95–108.
- 57. Riesch R, Martin RA, Langerhans RB (2013) Predation’s role in life-history evolution of a livebearing fish and a test of the Trexler-DeAngelis model of maternal provisioning. Am Nat181: 78–93.
- 58. Sheriff MJ, Love OP (2013) Determining the adaptive potential of maternal stress. Ecol Lett 16: 271–280.