香京julia种子在线播放

Aposematic animals have long been recognized to exhibit sluggish movements. Bates was the first to observe this, describing several strikingly colored butterflies as possessing “weak, slow flight” (Bates, 1862; Poulton, 1890). Wallace later wrote of the chemically defended, black-and-white striped skunk, “Its consciousness that it needs only to be seen to be avoided gives it that slowness of motion and fearlessness of aspect which are characteristic of most creatures so protected.” (Wallace, 1889). The phenomenon of sluggishness was later incorporated into the “chemical defense syndrome” (CDS), a general suite of characteristics commonly exhibited by chemically protected animals (Whitman et al., 1985). There are several hypotheses which seek to explain why aposematic animals exhibit these sluggish behaviors. One such hypothesis is that these slow movements evolved because they avoid triggering the attack responses of motion-oriented, ambush predators like frogs and praying mantids (Hatle and Faragher, 1998; Hatle et al., 2002). Another possibility is that these conspicuous, slow movements are themselves aposematic, acting as an additional signal of unprofitability to would-be predators. Expanding on Bates' early observations, a comparative study of butterfly species has quantitatively shown that distasteful species fly relatively more slowly than palatable species, indicating that sluggish flight can honestly signal prey defenses (i.e., “locomotor mimicry”; Chai and Srygley, 1990; Srygley, 1994, 2004). Another non-mutually exclusive hypothesis posits that these slow movements may make visual aposematic signals more apparent to predators, increasing the effectiveness of such signals (i.e., “raised exposure”; Speed et al., 2010).

We reserve “sluggishness” as a description of general locomotory movements, with “more sluggish” behavior indicating slower, seemingly careless movements, as it has traditionally been defined (Hatle and Faragher, 1998; Hatle and Whitman, 2001). However, animals can alter their movements in different contexts. When animals are actively pursued by predators they will often exhibit escape maneuvers. We define a seemingly careless demeanor in response to an immediate threat as “nonchalant.” “More nonchalant” indicates slower, less complex, delayed, and/or more rarely enacted escape responses in response to a potential predatory threat (Dowdy and Conner, 2016). Because of inherent costs of fleeing, prey are expected to balance perceived predation risk with the costs of fleeing such that they initiate escape at a distance which maximizes their fitness at the end of an encounter with a predator (i.e., “Optimal Escape Model”; Ydenberg and Dill, 1986; Cooper and Frederick, 2007). Flight initiation distance (FID) is known to vary depending on a variety of factors (e.g., body condition, distance to refugia) which likely influence prey assessment of predation risk and/or the costs associated with escape (Stankowich and Blumstein, 2005; Samia et al., 2016; Cooper, 2018). Chemical defenses may be one such factor influencing FID, however the escape responses of chemically defended prey have not been well-studied, and it is not clear whether sluggish animals should be expected to remain so when directly threatened. To our knowledge, the escape responses of aposematic animals in response to the threat of predation have only been investigated among dendrobatid frogs. Generally, these studies have found that aposematic frogs retain escape responses but delay fleeing relative to cryptic species, though variation in FID exists between aposematic species and based on the context of the encounter (Cooper et al., 2009a,b; Ozel and Stynoski, 2011; Dugas et al., 2015; Blanchette et al., 2017). Unfortunately, due to methodological limitations, the results of these studies confound escape decisions based on predation avoidance with trample avoidance (i.e., avoidance of being crushed), limiting the usefulness of their conclusions.

The bat-moth predator-prey system is an ideal system in which to study the escape behaviors of aposematic prey. Many moth species are equipped with ultrasound-sensitive ears which they use to detect bat predators. The ability of insects to detect the ultrasonic cries of bats led to their development of avoidance behaviors such as negative phonotaxis, spiraling erratic flight, and powered dives (Roeder, 1962). These maneuvers have been shown to be an effective means of dodging bat attacks (Acharya and Fenton, 1999; Triblehorn et al., 2008). Additionally, some moths possess secondary defenses which may alter their risk of predation. Many tiger moths (Erebidae: Arctiinae) sequester defensive toxins from their host plants and produce ultrasound in response to bat echolocation, advertising their unpalatability (i.e., “acoustic aposematism”; Acharya and Fenton, 1992; Dunning et al., 1992; Hristov and Conner, 2005; Barber et al., 2009; Dowdy and Conner, 2016). Unpalatability varies between species and may relate to the concentration and type of chemical compounds they have sequestered.

Interestingly, some tiger moths have been noted to lack any significant evasive flight response to bat attacks, even though they possess and utilize ultrasonic hearing (Goldman and Henson, 1977; Acharya and Fenton, 1992; Dunning et al., 1992; Dowdy and Conner, 2016). Field experiments with certain species of tiger moths have uncovered variation in escape behaviors in response to bat attacks (Dowdy and Conner, 2016). Two sympatric tiger moth species, Pygarctia roseicapitis and Cisthene martini, differed in both the likelihood of enacting evasive dives as well as in their unpalatability. The more unpalatable C. martini was significantly more nonchalant, rarely performing diving escape maneuvers. In contrast, the less unpalatable P. roseicapitis was less nonchalant, diving much more frequently when attacked by bats, indicating that a nonchalance continuum among aposematic species may exist.

In this study we examine the role that unpalatability plays in predicting the likelihood that individuals of a given species will enact escape behaviors in response to the threat of predation. Chemical defenses have not yet been recognized as a key factor affecting escape behaviors (Stankowich and Blumstein, 2005; Samia et al., 2016; Cooper, 2018). We hypothesized that more unpalatable tiger moth species would be less likely to exhibit escape maneuvers (i.e., more nonchalant) than their less unpalatable counterparts. This is the first study to comparatively examine the escape behaviors of aposematic insects in a natural context, compare the likelihood of escape between both aposematic and mimetic species, and construct a predictive model of escape behavior.

We found significant intraspecific variation in nonchalance and unpalatability (Figure 1). Statistical significance, level contrasts, and relative coefficients are given in Tables 1, 2. P. murina was significantly less nonchalant than all other species in our analysis. The prevalence of nonchalance between B. trigona and P. roseicapitis was not significantly different, though B. trigona was significantly less nonchalant as compared to C. arizonensis and C. martini. P. roseicapitis and C. arizonensis were not significantly different, however P. roseicapitis was significantly less nonchalant than C. martini. Finally, C. arizonenesis and C. martini were not significantly different.

Compared to all other species in our analysis, P. murina and B. trigona were significantly less unpalatable, whereas C. arizonensis and C. martini were significantly more unpalatable than all other species in our analysis. P. roseicapitis was significantly different from both of these groups, exhibiting an intermediate level of unpalatability.

Modeling nonchalance from unpalatability yielded a significantly strong, positive relationship (p < 0.05, Adj. R²: 0.87; Table 3). However, because we could only sample 5 species, our 95% prediction intervals are large, limiting our current ability to confidently estimate nonchalance from unpalatability (Figure 2). A summary of the estimated nonchalance and unpalatability for each species, along with their 95% confidence intervals, can be found in Table 4.

Our results demonstrate that, at least among tiger moths, significant interspecific variation in the likelihood of enacting evasive maneuvers in response to bat echolocation exists. This phenomenon forms a nonchalance continuum, with some species enacting escape maneuvers often, some rarely, and others lying somewhere between. In addition, we found significant interspecific differences in palatability among tiger moths. Variation in palatability has been noted in other insects, driven mainly by the host plants which they have fed on, and is generally known as a “palatability spectrum” (Brower et al., 1968, 1972; Turner, 1984). Further, we have shown that the position of tiger moth species along the nonchalance continuum and the palatability spectrum are strongly correlated. In other words, moths with more potent chemical defenses are more likely to display nonchalant flight. Interestingly, this is apparently independent of anti-bat strategy. Two species included in our analysis, B. trigona and C. arizonensis, produce clicks at very high rates and have been hypothesized or demonstrated to be capable of jamming bat sonar (Corcoran et al., 2010; Conner and Corcoran, 2012). Despite this, they each occupy significantly different places along the nonchalance continuum. However, consistent with our model, the large difference in unpalatability between these two species appears to be an important factor in determining their anti-bat evasive flight responses, with the more unpalatable C. arizonensis exhibiting nonchalant responses more frequently. Less unpalatable moths face an increased risk of consumption if captured by bat predators, and so they utilize evasive diving maneuvers more frequently, likely offsetting some of this risk. Additionally, more unpalatable prey may benefit from nonchalant flight by avoiding some of the potential costs of evasive maneuvers, including energy expenditure, opportunity costs related to hostplant- or mate-finding, or even increased exposure to terrestrial or aquatic predators (Guignion and Fullard, 2004; Yager, 2011). Our findings lend further support to an earlier study which demonstrated that palatability relates to evasive behaviors enacted from rest on substrates (e.g., flying away, dropping from vegetation) when exposed to tactile stimulation (Evans, 1983). These observations describe unpalatable prey exhibiting behaviors that could also be considered more nonchalant.

The relatively palatable P. murina produces anti-bat ultrasound with acoustic characteristics similar to those of its more unpalatable, sympatric congener P. roseicapitis (Dowdy and Conner, 2016; Dowdy, unpublished). This is similar to the relationship between Cycnia tenera and Euchaetes egle, two sympatric tiger moths native to eastern North America. Lab experiments with naïve bats have shown that the palatable E. egle was protected from predation by Eptesicus fuscus bats, acting as an acoustic Batesian mimic of C. tenera (Barber and Conner, 2007). It is likely that P. murina is an acoustic Batesian mimic of P. roseicapitis. It is intriguing then, that P. murina is less nonchalant than its more toxic model. It is possible that Batesian mimics will often “play it safe” by enacting escape responses more frequently than their more honestly signaling counterparts. This should be explored further with experiments confirming model-mimic relationships followed by a comparison of the frequency of evasive behaviors between members of these groups.

Of course, tiger moths are not the only chemically protected insects that employ evasive flight to contend with bat predation. Ears tuned to ultrasonic frequencies can be found in at least 5 orders of insects (i.e., Coleoptera, Lepidoptera, Mantodea, Neuroptera, and Orthoptera) and have evolved independently at least 14 times (Dawson et al., 2004; Greenfield, 2016). All 5 orders of ultrasound-sensitive insects have been reported to exhibit evasive flight responses in response to bat echolocation (Roeder, 1962; Miller and Oleson, 1979; Spangler, 1988; Yager et al., 1990; Dawson et al., 2004). Each of these orders, with the exception of Mantodea, contains examples of species utilizing a chemical defense of some kind (Eisner et al., 2005). Intraspecific variation in host plant chemical defenses as well as multiple host plant affiliations in some groups could generate intraspecific differences in chemical protection, potentially explaining some of the variation in nonchalance both among and within a variety of insect species, such as the recently reported nonchalant flight in fireflies in response to foraging bats (Leavell et al., 2018).

By gathering more evasive flight and palatability data for more species, we could also better predict nonchalance from palatability data. This is useful, as measuring palatability can be much simpler than making detailed observations of evasive flight. Additionally, if palatability could be confidently inferred from preserved museum specimens using analytical chemistry methods such as mass spectroscopy, then these traits could be determined on an unprecedented scale (Bowers, 2009; Anderson et al., 2016). Gaining a better understanding of the variation in nonchalance will help us better characterize how animals balance the potential costs of escape maneuvers with the potential benefits gained from eluding predators.

The bat-moth predator-prey system is a useful model for examining the escape behaviors of prey, both aposematic and non-aposematic. The sensory system that moths use to detect bat predators and their acoustic aposematic signals can be ablated by simple surgical impairment without compromising their overall physical integrity, their remaining behavioral repertoire, nor other cues produced by the specimen. Additionally, moth unpalatability can be modulated by rearing individuals on diets that are lacking defensive compounds. With this approach, the nonchalance, unpalatability, and/or aposematic signals of individual moths can be experimentally manipulated to yield greater insights into the costs, benefits, and factors influencing escape behavior. We believe the bat-moth predator-prey system is a promising new modality for understanding how the escape behaviors of prey are enacted and evolve.

The datasets analyzed for this study and the associated R scripts can be found in the FigShare repositories: Evasive Flight Data (10.6084/m9.figshare.9705788), Palatability Data (10.6084/m9.figshare.9706187), R Script (10.6084/m9.figshare.9706280), Figures and Tables (10.6084/m9.figshare.10318622), Tables in LaTeX format (10.6084/m9.figshare.10318604), and Supplemental Videos (10.6084/m9.figshare.10115180; 10.6084/m9.figshare.10115204; 10.6084/m9.figshare.10115207).

ND conceived and designed experiments and performed and analyzed experiments. ND and WC wrote the manuscript.