Generalist versus specialist strategies of plasticity: snail responses to predators with different foraging modes

  • Published on

  • View

  • Download


  • Generalist versus specialist strategies of plasticity: snailresponses to predators with different foraging modes


    *Department of Forestry and Natural Resources, Purdue University, West Lafayette, IN, U.S.A.Department of Biological Sciences, University of Pittsburgh, Pittsburgh, PA, U.S.A.


    1. Phenotypic plasticity is a common adaptation to environmental heterogeneity, and theory predicts

    that the evolution of constitutive versus plastic strategies should depend on the frequency of alterna-

    tive environments, the magnitude of constraints and the costs of plasticity per se. However, it is

    unclear how species should evolve when they experience more than two environments that favour

    divergent phenotypes, particularly when they have absolute constraints on their morphology.

    2. We examined the plasticity of three freshwater snail species (Helisoma anceps, H. campanulata and

    H. trivolvis) in response to three environments: (i) no predator; (ii) shell-invading water bugs (Belos-

    toma flumineum) and (iii) shell-crushing crayfish (Orconectes rusticus). We found distinct responses by

    each snail species to the predator treatments. Helisoma anceps starts with a relatively low, narrow and

    thick shell that becomes lower and thicker in response to crayfish but is unresponsive to water bugs.

    In contrast, H. campanulata starts with a relatively high, wide and thin shell that becomes lower and

    wider in response to water bugs but is unresponsive to crayfish. Helisoma trivolvis starts with a shell

    of intermediate height and width while the predators induce defences in different directions.

    3. These results suggest that H. trivolvis has a generalist plastic strategy while H. anceps and H. cam-

    panulata have specialised plastic strategies orientated against a single type of predator at the potential

    cost of being unable to respond to others.

    4. We then performed predation trials to determine predator preferences using a mixture of the three

    species. After 2 weeks of exposure to crayfish cues, H. anceps had higher survival than both H. trivol-

    vis and H. campanulata with uncaged crayfish. After 2 weeks of exposure to water bug cues, both

    H. trivolvis and H. campanulata had higher survival than H. anceps with uncaged water bugs. When

    predation trials were conducted after 5 weeks of exposure to predator cues, H. trivolvis and H. cam-

    panulata reached a size refuge from both predators and this shifted predation pressure to H. anceps.

    5. Collectively, these results suggest that closely related prey species with different absolute con-

    straints in their morphology had different defences that are either specialised or generalised to alter-

    native environments.

    Keywords: functional tradeoff, gastropod, inducible defence, phylogeny, selection


    Natural selection in heterogeneous environments may

    lead to the evolution of phenotypic plasticity, defined as

    the ability of a single genotype to produce different

    phenotypes in response to different environments

    (Schlichting & Pigliucci, 1998; Pigliucci, 2001).

    Phenotypic plasticity exists in many species and in

    response to a wide range of environmental conditions,

    and substantial variation in the expression of phenotypic

    plasticity can exist among closely related species

    (Harvell, 1991; Kusch, 1993; Colbourne, Hebert & Taylor,

    1997; Van Buskirk, 2002; Berendonk, Barraclough &

    Barraclough, 2003). However, the evolution of plasticity

    may be constrained by several mechanisms. For

    example, absolute constraints (sensu Brakefield, 2006)

    arise because the basic body plans of species are quite

    difficult to change by natural selection (i.e. new traits

    Correspondence: Jason T. Hoverman, Department of Forestry and Natural Resources, Purdue University, West Lafayette, IN 47907, U.S.A.


    2014 John Wiley & Sons Ltd 1

    Freshwater Biology (2014) doi:10.1111/fwb.12332

  • that break the constraint may be deleterious). Addition-

    ally, allocation tradeoffs can occur when resource limita-

    tion constrains simultaneous investment in multiple

    traits (Dewitt, 1998; Auld, Agrawal & Relyea, 2010).

    These constraints, combined with the amount of envi-

    ronmental heterogeneity experienced by a species over

    time or space, can drive the evolution of different mean

    phenotypes in closely related species (i.e. averaged

    across all environments), as well as different directions

    and magnitudes of phenotypic plasticity (Van Tienderen,

    1991; Dewitt, Sih & Wilson, 1998; Schlichting & Pigliucci,

    1998; Pigliucci, 2001; Van Kleunen & Fischer, 2007; Auld

    et al., 2010). Because closely related species are likely to

    be similar in biochemical, physiological and structural

    constraints, comparative studies have the potential to

    identify a core set of constraints that may limit the

    expression of environmentally induced traits, which

    may bias evolution towards fixed solutions to environ-

    mental heterogeneity (Pfennig et al., 2010).

    The inducible defences of freshwater snails represent

    an ideal system to assess how absolute constraints and

    allocation tradeoffs may influence the divergent evolu-

    tion of plasticity across species. For example, shell geom-

    etry creates constraints on shell shape. The shells of

    most gastropods can be described using three parame-

    ters: expansion rate (W), translation (T) and distance (D)

    of the generating curve from the axis of coiling (Raup,

    1962; Rice, 1998). Physical relationships among these

    shell parameters limit shell shape (i.e. not all regions of

    morphospace can be achieved) and therefore affect the

    range of options for morphological defences against var-

    ious predators. For example, species with rapid shell

    expansion rates (i.e. a coiling tube that rapidly increases

    in diameter) produce shells with relatively large aper-

    tures, which are more vulnerable to predators that enter

    the shell. However, these species can invest more in

    shell thickness, a defence against shell-crushing preda-

    tors, because fewer coils around the central axis are

    needed to increase overall body size (Raup, 1962). Alter-

    natively, snails with slow expansion rates generate shells

    with relatively small apertures that are difficult to enter,

    although such shells are typically thinner because more

    coils around the central coiling axis are required to grow

    to a particular body size (Raup, 1962). In addition to

    absolute constraints on shell geometry, snails also

    face allocation tradeoffs (Dewitt & Langerhans, 2003;

    Hoverman & Relyea, 2007a, 2009); for instance, there can

    be a tradeoff between investing a limited amount of

    shell material to thickness or coiling (Russell-Hunter,

    1978; Kemp & Bertness, 1984; Brodersen & Madsen,

    2003). As a result, snails commonly face tradeoffs in

    how they respond to shell-invading versus shell-crush-

    ing predators (Dewitt, 1998; Hoverman & Relyea, 2008,

    2009; Bourdeau, 2009). Thus, interspecific differences in

    expansion rate, coupled with constraints on shell thick-

    ness, could influence patterns of phenotypic plasticity

    and phenotypic diversification of snail species (Edgell &

    Miyashita, 2009).

    We examined the inducible defences of three closely

    related planorbid snails (Helisoma trivolvis, H. anceps and

    H. campanulata; Fig. 1) to determine how these snails

    respond to predators with different foraging modes,

    given the absolute constraints and allocation tradeoffs

    that limit shell morphology. The three snail species

    occur together in semipermanent to permanent water-

    bodies, where they encounter a diversity of predators

    including water bugs, crayfish and fish (Hoverman et al.,

    2011). In a series of studies, we have explored the

    responses of H. trivolvis to different predators (Hover-

    man, Auld & Relyea, 2005; Hoverman & Relyea, 2007a,b,

    2008, 2009). In the presence of the water bug Belostoma

    flumineum, H. trivolvis invests in shell coiling but the

    aperture remains relatively small because of their

    moderate expansion rate; this reduces the ability of the

    bug to reach the snails soft tissues when withdrawn

    inside the shell. In contrast, H. trivolvis forms thicker

    shells in the presence of the crayfish Orconectes rusticus,

    which reduces the predators ability to crack or crush

    the shell.

    In contrast to the extensive research on H. trivolvis,

    there appear to be no studies on the predator-induced

    morphology of H. campanulata and H. anceps. Among the

    three species, the main difference in shell shape is the

    rate of shell expansion (Raup, 1962). Visual inspection

    suggests that the expansion rate is low for H. campanula-

    Fig. 1 Left side (i.e. spire) view of Helisoma anceps, H. campanulata

    and H. trivolvis. These three species differ in the rate at which the

    diameter of the shell increases with each rotation around the

    coiling axis (termed w; Raup, 1962); w is low for H. campanulata,

    intermediate for H. trivolvis and high for H. anceps, respectively. As

    a consequence of variation in w, H. anceps exhibits a relatively large

    aperture for a given body mass, whereas H. campanulata exhibits a

    relatively small aperture.

    2014 John Wiley & Sons Ltd, Freshwater Biology, doi: 10.1111/fwb.12332

    2 J. T. Hoverman et al.

  • ta, intermediate for H. trivolvis and high for H. anceps.

    Consequently, H. campanulata has a relatively small shell

    aperture relative to its overall size, whereas H. trivolvis

    has an intermediate aperture and H. anceps has a rela-

    tively large aperture. Additionally, shell thickness also

    appears to differ among species; shells are relatively

    thick for H. anceps, intermediate for H. trivolvis, and rela-

    tively thin for H. campanulata (Osenberg & Mittelbach,

    1989; Brown, 1998).

    These differences in basic shell geometry and thick-

    ness might constrain the predator-induced defences of

    each species, resulting in differences in predation risk by

    shell-invading versus shell-crushing predators. For

    example, because H. anceps has a thick shell and a high

    shell expansion rate, it should be well defended against

    shell-crushing predators but vulnerable to shell-invading

    predators. Because H. campanulata has a thin shell with a

    low shell expansion rate, it should be well defended

    against shell-invading predators but vulnerable to crush-

    ing predators. To test these hypotheses experimentally,

    we assessed the phenotypic responses of the snails to

    water bugs or crayfish and then examined the relative

    susceptibility of each snail species to the predators.


    Induction experiment

    The goal of the induction experiment was to assess

    predator-induced morphology in the three Helisoma spe-

    cies. We collected ~100 adults of H. trivolvis, H. anceps

    and H. campanulata from ponds near the University of

    Pittsburghs Pymatuning Laboratory of Ecology (PLE) in

    Linesville, PA. These ponds contain both water bugs

    (Belostoma flumineum) and crayfish (Orconectes rusticus).

    For each snail species, 10 individuals were placed into

    each of 10 culture pools filled with 100 L of well water.

    Egg deposition began immediately and continued until

    the adults were removed after 2 weeks. Upon hatching,

    snails were fed rabbit food ad libitum until the experi-

    ment began.

    In a mesocosm experiment, we examined the effects of

    caged predators on the growth and morphology of each

    species. We designed a completely randomised, factorial

    experiment composed of three predator treatments (no

    predator, caged water bug [B. flumineum], or caged cray-

    fish [O. rusticus]) crossed with the three snail species.

    These nine treatments were replicated eight times for a

    total of 72 experimental units.

    The experimental units were 90-L pools filled with

    well water. We added 10 g of rabbit food and an aliquot

    of pond water, containing periphyton, phytoplankton

    and zooplankton, to sustain food for snails and maintain

    water quality. Each pool received 100 juvenile snails of

    the appropriate species. Initial mean mass 1 SD ofH. trivolvis, H. anceps or H. campanulata was 2.0 1.4 mg, 1.2 0.6 mg and 1.5 0.9 mg, respectively. Foreach snail species, 20 snails were set aside to assess mor-

    tality due to handling; 24-h survival was 100%.

    After adding the snails, we placed a single predator

    cage into each pool. The cages were made from corru-

    gated pipes (10 cm long 9 10 cm diameter) capped with

    shade cloth. For caged predator treatments, we added

    one water bug or crayfish to each cage. Caged predators

    emit water-borne chemical cues, which provide the

    opportunity for prey to detect and respond to predators

    without reducing prey density (Chivers & Smith, 1998).

    The caged predators were fed 300 mg of snail biomass

    (total wetmass including shells, two to five snails) of the

    appropriate snail species three times per week. Based on

    previous research, this amount of consumed snail bio-

    mass by the predators is sufficient to elicit phenotypic

    responses in H. trivolvis (Hoverman & Relyea, 2007b,

    2008, 2009). The predators consumed all the snails

    between feedings. To equalise disturbance, we briefly

    lifted the cages in the no-predator treatment from the

    water and then returned them. We placed a shade cloth

    lid over each pool to prevent colonisation by insects and


    During the experiment, we observed that the spe-

    cies varied greatly in growth rates; growth was fast in

    H. trivolvis, intermediate in H. campanulata and slow in

    H. anceps. Because analyses of morphological plasticity

    are sensitive to differences in mass, we decided to take

    down the experimental units for each species at different

    times so that they were similar in mass at the end of the

    experiment. Although this approach resulted in differ-

    ences among the species in duration of predator expo-

    sure, our previous work with H. trivolvis found that the

    magnitude of predator-induced morphological change is

    relatively constant over ontogeny (Hoverman & Relyea,

    2007a, 2009). For H. trivolvis, H. campanulata and

    H. anceps, the experiment was ended after 14, 21 and

    39 days, respectively. On each date, all surviving snails

    were removed and preserved in 10% formalin. In one

    experimental unit, all the H. campanulata died and this

    was excluded from analyses. For the remaining experi-

    mental units, survival was high (>95%) and did not

    differ among caged predator treatments or snail species

    (predator, F2,63 = 0.7, P = 0.499; species, F2,63 = 0.3,

    P = 0.706; interaction, F4,63 = 1.3, P = 0.274). For each

    experimental unit, 10 individuals were randomly

    2014 John Wiley & Sons Ltd, Freshwater Biology, doi: 10.1111/fwb.12332

    Defensive strategies of Helisoma snails 3

  • selected and dried at 60 C for 24 h. Each individual

    was then weighed to the nearest mg (total dry mass

    included shell and tissue) and measured for shell width

    and height, and aperture width and height using digital

    imaging software (Optimas Co., Bothell, WA, U.S.A.).

    We also measured the shell thickness of each snail at the

    leading edge of the aperture using digital calipers.

    To examine the effects of our caged predator treat-

    ments on snail morphology, we began by assessing the

    allometric relatio...


View more >