Evolution was once thought to work exclusively as a gradual process (Darwin, 1859), where speciation followed the accumulation of divergent traits primarily in isolated populations (Carroll et al., 2007). Yet decades of research have transformed this model of evolution into one that emphasizes the importance of behavior (Lapiedra et al., 2018; Grant and Grant, 2010), while accommodating not only rapid phylogenetic and phenotypic diversification (Rudh et al., 2011; Uy and Borgia, 2000; Kettlewell, 1956), but also speciation with gene flow (Servedio, 2016; Servedio, 2004; Kopp et al., 2018). Theory predicts that such instances of rapid speciation may occur provided there is sufficient phenotypic disparity between diverging population to erect reproductive barriers, but this implies that other factors that give rise to phenotypic variation also play an important role in this process. One of these factors is an organism’s physiology, which provides the functional architecture that defines how phenotypes can evolve and/or emerge anew. However, studies rarely explore the role of certain physiological systems in the process of sympatric speciation, creating a gap in our understanding of modern evolutionary biology.

A major way that physiology influences evolution is through constraint, due to numerous trade-offs between systems or processes that function in a mutually exclusive manner (Roff and Fairbairn, 2007; Alexander, 1985). Skeletal muscle performance is a prime example of this phenomenon: its design entails performance traits that are inversely related to each other, such as between speed and endurance (Vanhooydonck et al., 2014; Wilson et al., 2002). At the same time, skeletal muscle also actuates many sexually-selected behaviors that act as the basis for mate choice (Goller and Suthers, 1996; Garcia et al., 2012; Fuxjager et al., 2017a). This suggests that performance trade-offs inherent to muscle itself can impede the formation of reproductive barriers, particularly in cases where behavioral divergence should mediate speciation. Evaluating this idea requires a simultaneous examination of a species’ evolutionary history, as well as whole-organism and muscle performance. Few studies unite all three perspectives, however, leaving an unclear relationship between animal design, behavior, and evolutionary process.

Here, we address this issue by exploring how performance trade-offs borne out via skeletal muscle shaped the evolution of sexual display behavior during a rapid speciation event. We study a small genus of Neotropical birds called bearded manakins (Aves: Pipridae). This group last shared a common ancestor roughly 300,000 years ago (Brumfield et al., 2008; Tello et al., 2009), before splitting into four species: white-bearded manakins (M. manacus), golden-collared manakins (Manacus vitellinus), white-collared manakins (M. candei), and orange-collared manakins (M. aurantiacus). We specifically examine how disruptive selection operates on one of the bird’s main courtship signals—the roll-snap—to generate phenotypic divergence sufficient to facilitate assortative mating (Servedio, 2016). Males produce this behavior by rapidly hitting their wings together over their back, generating a mechanical trill-like sound (Fuxjager et al., 2017a). The result is a repetitive rhythmic signal, which is a common signal type in animal courtship (Mowles and Ord, 2012). These displays convey information through multiple component signals, but one of particular importance in reshaping a receiver’s behavior is signal speed (Schuppe and Fuxjager, 2018; Ord et al., 2007; Podos, 2001). As such, we hypothesize that speed is an important feature of roll-snap performance. Although the degree to which roll-snap speed itself is sexually selected remains unknown, previous studies in golden-collared manakins show that females discriminate millisecond-scale differences in other displays and prefer to mate with males that display faster (Barske et al., 2011). Therefore, the speed with which a male hits his wings together likely functions as an important courtship signal that is subject to strong sexual selection. If so, then small changes in female preferences for speed may provide the selection pressure necessary for display divergence. However, disruptive selection may not be able to freely drive sister species to exhibit opposite patterns of directional selection for this trait, largely because the speed with which skeletal muscles control wing movements is constrained by important performance trade-offs, such as one between speed and endurance (Vanhooydonck et al., 2014; Wilson et al., 2002). This leads to the question: what role did intrinsic muscle performance trade-offs play in the rapid diversification of these species?

As a first step toward answering the question posed above, we evaluated whether speciation of bearded manakins occurred under sympatry, a special case of speciation with gene flow (Kopp et al., 2018). We accomplished this using phylogeographic models (O'Donovan et al., 2018; Lemmon and Lemmon, 2008; Pagel et al., 2004) to reconstruct the ranges of ancestral Manacus populations (Figure 1a). The model output (Figure 1b) shows that the bearded manakins’ most recent common ancestor occupied a geographic range similar to that of the western population of extant white-bearded manakins, giving rise to this species before migrating west across the Isthmus of Panama (Video 1). Golden-collared manakins were the next to split off, emerging in a geographic range that overlapped between 49% and 73% (95% credible intervals) with that of the final common ancestor to both white-collared and orange-collared manakins. This wide-ranging ancestral overlap is therefore consistent with speciation occurring in sympatry, typically considered to occur when ranges overlap ≥25% (Seddon, 2005; Cooney et al., 2017; Tobias et al., 2014). Furthermore, because modern bearded manakins are mostly allopatric (Figure 1c), geographic isolation of these birds likely followed their diversification, rather than preceding it. This is consistent with previous work in golden-collared and white-collared manakins, which hybridize readily in a small region of range overlap in Panama (Parsons et al., 1993). In the hybrid zone, directional selection for plumage ornaments is strong enough for golden-collared manakin traits to cross the hybrid zone and introgress into the white-collared manakin population (Parsons et al., 1993; McDonald et al., 2001; Brumfield et al., 2001). Moreover, whole-genome analysis of the two species finds high rates of introgression and genetic differentiation at a wide range of unrelated loci, which is consistent with an evolutionary history characterized by reproductive isolation following adaptive divergence in various traits (Parchman et al., 2013). Thus, mechanisms other than those linked to geographic isolation likely help drive rapid speciation in the genus.

Figure 1 with 1 supplement see all

Download asset Open asset

Video 1

Download asset

If courtship display divergence played a role in speciation, then species arising with gene flow should exhibit different display phenotypes. We therefore examined species differences in the acrobatic roll-snap display by analyzing acoustic recordings of roll-snaps for each species (Table 1), measuring both signal speed (snaps s⁻¹) and length (total number of snaps). Speed differed significantly among the species (Figure 2a; F_3,146=9.203, p<0.0001), with golden-collared manakins producing the fastest roll-snaps, white-bearded and white-collared manakins (no ancestral overlap) both producing intermediate speed roll-snaps, and orange-collared manakins producing the slowest roll-snaps. The magnitude of mean differences is relatively small overall (<5 Hz), but also reflects larger differences in the upper range of display speed. For example, the fastest roll-snap observed in white-collared manakins (61 Hz) falls within the interquartile range of the golden-collared manakin’s speed distribution [55.9,61.1]. Because phenotypic constraints typically impact trait evolution by restricting the extremes of a distribution (Wilkins et al., 2013), this suggests that shifting constraint space may underlie differences in speed. At the same time, we found that roll snap length in all four species was statistically indistinguishable (Figure 2b; F_3,153=0.229, p=0.876). Therefore, the roll-snap diverged among species exclusively in terms of signal speed (i.e. snap repetition rate), but not in terms of the total number of snaps. Therefore, if females attend to differences in roll-snap speed as they do with other signals (Barske et al., 2011), conspecific mate preference should be reinforced by divergent roll-snap speeds (McDonald, 1989). Our data are thus consistent with a model in which phenotypic divergence among sympatric populations plays a role in supporting assortative mating (Servedio, 2016; Kopp et al., 2018).

Figure 2

Download asset Open asset

Figure 2—source data 1

Data_acoustics.txt: Measurements of roll-snap speed and length obtained from audio recordings of displaying birds.

Includes accession numbers for locating the original recordings on internet archives.

https://doi.org/10.7554/eLife.40630.006

Download elife-40630-fig2-data1-v1.txt

Evolutionarily modifying a signal to become faster can sometimes incur costs to other performance attributes, such as endurance (Van Damme et al., 2002; Vanhooydonck et al., 2001; Reidy et al., 2000, c.f. Sorci et al., 1995). Divergence in the roll-snap may therefore be constrained by these trade-offs, especially in species that evolve faster—but equally as long—roll-snaps. To examine whether this evolutionary impasse is present, we used quantile regression to test for an inverse relationship between the speed and length of each species roll-snap (Cade and Noon, 2003; Koenker and Machado, 1999; Miles et al., 2018; Wilson et al., 2014). The mere existence of such a relationship would indicate that a given species was signaling near (or at) its phenotypic limit with respect to these measures of performance. Indeed, at the uppermost testable quantile (τ = 0.9), we detected this trade-off in both golden-collared (Figure 3a; t = -4.03, p<0.001) and white-collared manakins (Figure 3b; t = -2.76, p=0.007), but not in white-bearded (Figure 3c; t = -0.36, p=0.722) and orange-collared manakins (Figure 3d; t = -0.70 p=0.490). Roll-snaps are therefore differentially constrained across species.

Figure 3 with 1 supplement see all

Download asset Open asset

Even among two species undergoing a performance constraint between speed and endurance, there can be variation in terms of the trade-off’s steepness (slope), as well as the proportion of individuals in the population that encounter it. If the trade-offs we uncover are the result of endurance costs, then the negative linear relationship between speed and length should be steeper in the species with the faster display (i.e. golden-collared manakins, see Figure 2a). This was indeed the case (F_1,230=22.1, p=0.001), supporting the notion that performance costs to speed become greater as selection drives the evolution of the signal further into the constrained trait-space. Similarly, because golden-collared manakins have a faster display overall than white-collared manakins (post-hoc: z = 2.5, p=0.038), more individuals should produce displays fast enough to encounter endurance constraints. Our data again support this idea, as nearly all golden-collared manakins (top 80% of the population) experience a cost to display speed (Table 2), given the signal’s length (Figure 3a; at τ = 0.2: t = -2.95, p=0.004), whereas only the top 20% of white-collared manakins display fast enough to experience such a cost (Figure 3b; at τ = 0.8: t = -2.33, p=0.028). Thus, for these two species, the rate at which an individual snaps its wings together to display is influenced by the inverse relationship between speed and endurance, with the effect of this constraint appearing greater in golden-collared manakins. This may be a consequence of the fact that golden-collared manakins have a faster roll-snap than the other species, but without a change in overall display length. Their roll-snap speed appears to have diverged just enough to maintain the signal’s length, with trade-offs consequentially appearing at the level of the individual.

The origin of behavioral trade-offs may lie in the physiological and/or morphological underpinnings of display production (Podos, 2001; Derryberry et al., 2012). For the roll-snap, skeletal muscle performance may be one of the limiting factors, considering that muscle tissue itself is thought to be the origin of the speed-endurance trade-off (Vanhooydonck et al., 2014; Wilson et al., 2002; Bottinelli et al., 1994). In bearded manakins, the scapulohumeralis caudalis muscle (SH, Figure 4a) is the primary actuator of the roll-snap—it contracts when a male’s wings are raised vertically above its back, causing the wrists to collide (snap) over the axial mid-line (Fuxjager et al., 2017a). Otherwise, the SH is a flight muscle, acting as a ‘strut’ during powered locomotion by rotating and retracting the wing (Dial, 1992). Past work shows that bearded manakins maintain an extremely fast SH, likely due to selection for a fast display (Barske et al., 2011; Fuxjager et al., 2016a; Fuxjager et al., 2017b); thus, one might expect that the evolution of SH speed contends with a decrease in endurance. To test this idea, we quantified muscle performance first in the golden-collared manakin (the fastest species) and then in the white-collared manakin. Although the white-collared manakin was not the species with the slowest roll-snap, we chose it as a basis of comparison due to the well-studied evolutionary history between these two populations. White-collared and golden-collared manakins are the only pair of Manacus species that are well known to hybridize today (McDonald et al., 2001), which means that signal divergence is likely important to mediating reproductive barriers in the present day.

Figure 4

Download asset Open asset

Figure 4—source data 1

Data_fatigue.txt: Contractile phase measurements collected from each twitch cycle during stimulation of golden-collared and white-collared manakin SH muscles.

https://doi.org/10.7554/eLife.40630.012

Download elife-40630-fig4-data1-v1.txt

We used in situ recordings of SH twitch speed (Fuxjager et al., 2016a; Fuxjager et al., 2017b) to test whether skeletal muscle performance underlies differences in display behavior. Accordingly, we measured the percent relaxation of the SH at different stimulation frequencies (30–100 Hz), modeling relative relaxation (0–100%) as a function of muscle stimulation speed (Figure 4b). Our results confirmed previous work (Fuxjager et al., 2016a) by indicating that golden-collared manakins maintain an extremely fast SH—that is, this muscle relaxes more quickly at high frequencies than it does in white-collared manakins (Figure 4c; model difference: F_3,49=15.2, p<0.001). The muscle therefore also maintains a significantly greater half-relaxation frequency in the golden-collared manakin, which is the maximum frequency at which the muscle contracts and relaxes at 50% of its predicted functional range (Figure 4d; F_1,49=79.8, p<0.001). Next, to determine if the ability of the SH to perform at such speeds is limited by endurance costs, we modeled the tissue’s percent relaxation as a function of the stimulation pulse number (1-8) at each given stimulation frequency. This reflects how the muscle’s performance changes at different speeds in response to repeated stimulations. We found that the golden-collared manakin SH relaxed between 75% and 100% when stimulated at 50, 60 and 70 Hz (Figure 5a). However, when stimulated at ≥80 Hz, its percent relaxation significantly declined with each twitch (Figure 5a and Table 3; F_1,12=18.7, p=0.005 at 80 Hz; F_1,12=33.9, p<0.001 at 90 Hz; F_1,12=22.8, p=0.003 at 100 Hz). Indeed, we find that the level of relaxation dropped from around 75–100% during the first contraction to around 20–50% as the stimulation train progressed. By contrast, in the white-collared manakin, percent relaxation of the SH never changed across any of the stimulation trains (Figure 5b and Table 3). Instead, only overall percent relaxation decreased with increasing frequency (see above, Figure 4c). These data therefore suggest that the golden-collared manakin SH experiences a form of ‘rapid fatigue’ when stimulated at especially fast frequencies. This performance attribute contrasts with fatigue as it is normally discussed, where the latter is typically defined as a decrease in speed and force production over periods of time greater than the fractions-of-a-second that we show herein (Wilson et al., 2002; Allen et al., 2008). Nonetheless, we expect that the ability to resist this ‘rapid fatigue’ similarly qualifies as a form of muscle endurance per se (the ability to resist fatigue; Wilson et al., 2002).

Figure 5 with 1 supplement see all

Download asset Open asset

To further examine how this divergence in SH performance relates to display behavior, we modeled the way that the observed ‘rapid fatigue’ changed in response to stimulation frequency. As such, we first computed the slopes that describe the relationships between percent relaxation and stimulus number within the different stimulation trains (see above; Figure 5a and b). We then plotted these values as a function of overall stimulation frequency, finding that the SH of the golden-collared manakin experiences a progressively greater decline in performance when stimulated at high speeds (Figure 5c; F_1,6=27.7, R² = 0.82, p=0.002). This effect is absent in white-collared manakins (Figure 5c; F_1,7=0.15, R² = 0.02, p=0.710), which instead maintain a constant (fused) state when subjected to the same fast stimulation frequencies. Importantly, for the golden-collared manakin, we show that the upper confidence limit of this regression line crosses 0 at ≈60 Hz—a value that represents the relative speed at which the trade-off with endurance should begin to manifest. This roughly correspond to the species’ average roll-snap speed, which is 58.4 Hz (Figures 2a and 5c). Thus, sexual selection in golden-collared manakins has likely driven the evolution of the roll-snap to its maximal speed, such that it is produced at the SH’s performance limit due to the tissue’s intrinsic trade-off with endurance. Further support for this idea comes from a subsequent analysis, in which we found that male golden-collared manakin produce significantly shorter roll-snaps when they are at or above 60 Hz, compared to less than 60 Hz (β = −4.32, F_1,119=46.7, p<0.001; Figure 5—figure supplement 1).

What explains the rapid decline in relaxation with multiple contractions at high speed? Distinct mechanisms may shape the different phases of the contraction-relaxation cycle; thus, we quantified shortening and lengthening periods in twitch recordings that were both (i) right above each species’ average roll-snap speeds (golden-collared manakin threshold = 60 Hz, white-collared manakin threshold = 58 Hz) and (ii) at a high frequency stimulation (80 Hz) (Figure 6). Both measures showed considerable variation, but for both species we found no significant changes within the twitch series (p_min = 0.20; Figure 6). This indicates that shortening and lengthening duration remained consistent during the stimulation trains. Nonetheless, in the golden-collared manakin, we know that percent relaxation declines at high stimulation frequencies in this same timeframe. Thus, the overall ‘rapid fatigue’ response is likely a result of decreased capacity for relaxation, pointing to processes underlying this event as a potential limiter of muscle performance. This may be linked to constraints on calcium buffering and/or re-uptake mechanisms, given they play an important role in setting muscle relaxation rates (Syme and Josephson, 2002; Rome et al., 1996). Regardless of which cellular mechanism ultimately limits SH performance in the golden-collared manakin, this is consistent with the fundamental framework of selection contending with constraint—the phenotype favored by selection (in this case, display speed) will evolve following directional selection until it is impeded by the underlying rate-limiting mechanism. In this case, it appears that the golden-collared manakin SH has been modified to allow the muscle to sustain rapid contraction-relaxation cycling, but different components of myocytic machinery have likely evolved in a mismatched fashion.

Figure 6

Download asset Open asset

Figure 6—source data 1

Data_twitchspeed.txt: Twitch speed measurements collected from muscle recordings of golden-collared and white-collared manakin SH muscles.

https://doi.org/10.7554/eLife.40630.016

Download elife-40630-fig6-data1-v1.txt

At the cellular level, muscle relaxation requires both rapid cross-bridge detachment and swift calcium cycling (Syme and Josephson, 2002; Rome et al., 1996). Past work in the golden-collared manakins suggests that the SH can achieve the former, with measures of maximal shortening velocity surpassing the maximum roll-snap speed (Fuxjager et al., 2016a). Thus, the factor that limits SH speed for more than two or three twitches—if not related to cross-bridge detachment rates—is likely related to processes governing calcium cycling within the myocyte (Syme and Josephson, 2002). This is indicative of evolutionary discordance between different facets of the muscle’s contractile machinery, which ultimately put the brakes on sexual selection for a rapid display. Further evolutionary elaboration of the roll-snap speed would likely require modifications to the calcium handling properties, alongside other adaptations for the performance of this signal in bearded manakins (Friscia et al., 2016; Fuxjager et al., 2016b; Fuxjager et al., 2016c). Notably, limitations to muscular calcium handling are not typically associated with the speed-endurance trade-off, which is often considered to result from muscle fiber type composition (Rivero et al., 1993; Komi, 1984; Esbjörnsson et al., 1993). This discrepancy is likely the result of differences between the so-called ‘rapid fatigue’ we observe and the more standard muscle fatigue that occurs when performance is depleted after prolonged muscle use.

General discussion

This study provides a blueprint that outlines how trade-offs in skeletal muscle performance influence a rapid sympatric speciation event. We use bearded manakins, which are Neotropical birds that separated into distinct species relatively recently. Our results show that the species in this complex evolved in a geographic scenario characterized by considerable range overlap, which means that that rapid speciation was likely facilitated by strong assortative mating (Servedio, 2016; Kopp et al., 2018). Bearded manakin divergence took place after ancestral populations migrated northwest out of Amazonia, across the Andes, and along the Isthmus of Panama. When the golden-collared manakin first diverged from its shared common ancestor with white- and orange-collared manakins, its range spanned the entire isthmus. As the species split, their roll-snap speeds also diverged for the three taxa that were ancestrally sympatric, and their ranges also began contracting to generate the (mostly) allopatric configuration seen today (Video 2). Specifically, we found that the golden-collared manakin’s roll-snap increased in speed, presumably at the same time as its SH was modified to support the especially rapid contraction-relaxation cycling that was necessarily to generate the movements that make up this behavior. The underlying driver for signal and range differentiation may have been shifts in female preferences for male courtship display performance (Kopp et al., 2018; Servedio and Burger, 2014). Alternatively, emergent differences in roll-snap speed may have supported conspecific mate preference independently of directional selection for speed. This may explain why orange-collared manakins, which are not known to hybridize with golden-collared manakins, uniquely exhibit a sharp decrease in roll-snap speed as their range moved south of the sympatric ancestral range. By contrast, the white-collared manakin (which did not undergo a large degree of speed differentiation) still interbreeds with golden-collared manakins where their ranges overlap. It also remains possible that other aspects of the roll-snap phenotype not assessed here—for example, the display’s amplitude or modulation thereof—may play a similar role in supporting divergence, but this remains to be seen.

Video 2

Download asset

During the parallel differentiation of species, plus their ranges and display phenotypes, the golden-collared manakin alone exhibited an increase in roll-snap speed. Such a change should incur some cost to the signal’s length, but we found no matching macroevolutionary shift in roll-snap length. Instead, a trade-off between speed and endurance is apparently borne out at the population level (rather than a macroevolutionary one), where only the species that evolved faster displays experience a trade-off between speed and length. We then show that this evolutionary pattern is explained by contraction-relaxation cycling kinetics in the skeletal muscle that actuates the roll-snap display—namely the SH. Thus, the species with the fastest display, the golden-collared manakin, also has the fastest SH. Further analysis of SH performance, however, reveals that this same bird also undergoes a decline in percent relaxation over repeated contractions at speeds greater than 60 Hz, which corresponds to the species’ average roll-snap speed. This helps explain why the SH appears to be faster than the bird’s actual display—that is, an intrinsic trade-off between muscle speed and endurance (length) blunts any potential effect of sexual selection for a faster display. Finally, we separately examined whether this endurance ‘cost’ to speed influences the duration of contraction and relaxation within a stimulation series, and we find that the timing of these two variables remain constant across multiple stimulations. Importantly, these effects occur despite the muscle showing an overall decline in percent relaxation, which points to the process of relaxation in and of itself as the origin of phenotypic constraint on roll-snap evolution.

Altogether, our data show how muscle physiology can influence the process by which divergent selection drives phenotypic evolution to support speciation with gene flow. Indeed, for the nascent radiation of tropical bearded manakins, this processes likely partially hinges on changes to an acrobatic mating dance that inextricably linked to the muscular control of behavioral outflow. Thus, if selection on this display is to proceed, it must contend with constraints imposed by this tissue’s design. We provide a glimpse into how this might occur, while also showing that additional changes are necessary to the muscle for selection on the golden-collared roll-snap to proceed further. Of course, other traits like plumage ornamentation have also diverged due to sexual selection (McDonald et al., 2001), although recent work suggest sexual dichromatism alone is not likely to facilitate species coexistence or speciation in sympatry (Cooney et al., 2017). Nonetheless, our study supports the idea that organism’s physiological design is critical for creating the landscape in which phenotypic evolution occurs, and therefore contributes to the process by which species diverge.

All data generated or analyzed during this study are included in the manuscript and supporting files.

1. 1. Alexander RM

   (1985) The ideal and the feasible: physical constraints on evolution

   Biological Journal of the Linnean Society 26:345–358.

   https://doi.org/10.1111/j.1095-8312.1985.tb02046.x

   • Google Scholar
2. 1. Allen DG
   2. Lamb GD
   3. Westerblad H

   (2008) Skeletal muscle fatigue: cellular mechanisms

   Physiological Reviews 88:287–332.

   https://doi.org/10.1152/physrev.00015.2007

   • PubMed
   • Google Scholar
3. 1. Barske J
   2. Schlinger BA
   3. Wikelski M
   4. Fusani L

   (2011) Female choice for male motor skills

   Proceedings of the Royal Society B: Biological Sciences 278:3523–3528.

   https://doi.org/10.1098/rspb.2011.0382

   • PubMed
   • Google Scholar
4. 1. Benjamini Y
   2. Hochberg Y

   (1995)

   Controlling the false discovery rate: a practical and powerful approach to multiple testing

   Journal of the Royal Statistical Society B 57:289–300.

   • Google Scholar
5. 1. Bottinelli R
   2. Betto R
   3. Schiaffino S
   4. Reggiani C

   (1994) Unloaded shortening velocity and myosin heavy chain and alkali light chain isoform composition in rat skeletal muscle fibres

   The Journal of Physiology 478:341–349.

   https://doi.org/10.1113/jphysiol.1994.sp020254

   • PubMed
   • Google Scholar
6. 1. Brumfield RT
   2. Jernigan RW
   3. McDonald DB
   4. Braun MJ

   (2001) Evolutionary implications of divergent clines in an avian (Manacus: aves) hybrid zone

   Evolution 55:2070–2087.

   https://doi.org/10.1111/j.0014-3820.2001.tb01322.x

   • Google Scholar
7. 1. Brumfield RT
   2. Liu L
   3. Lum DE
   4. Edwards SV

   (2008) Comparison of species tree methods for reconstructing the phylogeny of bearded manakins (Aves: Pipridae, Manacus) from multilocus sequence data

   Systematic Biology 57:719–731.

   https://doi.org/10.1080/10635150802422290

   • PubMed
   • Google Scholar
8. 1. Cade BS
   2. Noon BR

   (2003) A gentle introduction to quantile regression for ecologists

   Frontiers in Ecology and the Environment 1:412–420.

   https://doi.org/10.1890/1540-9295(2003)001[0412:AGITQR]2.0.CO;2

   • Google Scholar
9. 1. Carroll SP
   2. Hendry AP
   3. Reznick DN
   4. Fox CW

   (2007) Evolution on ecological time-scales

   Functional Ecology 21:387–393.

   https://doi.org/10.1111/j.1365-2435.2007.01289.x

   • Google Scholar
10. 1. Cooney CR
    2. Tobias JA
    3. Weir JT
    4. Botero CA
    5. Seddon N

    (2017) Sexual selection, speciation and constraints on geographical range overlap in birds

    Ecology Letters 20:863–871.

    https://doi.org/10.1111/ele.12780

    • PubMed
    • Google Scholar
11. Book

    1. Darwin C

    (1859)

    The Origin of Species by Means of Natural Selection: Or, the Preservation of Favored Races in the Struggle for Life

    Oxford: Oxford University Press.

    • Google Scholar
12. 1. Derryberry EP
    2. Seddon N
    3. Claramunt S
    4. Tobias JA
    5. Baker A
    6. Aleixo A
    7. Brumfield RT

    (2012) Correlated evolution of beak morphology and song in the neotropical woodcreeper radiation

    Evolution 66:2784–2797.

    https://doi.org/10.1111/j.1558-5646.2012.01642.x

    • Google Scholar
13. 1. Dial KP

    (1992) Activity patterns of the wing muscles of the pigeon (Columba livia) during different modes of flight

    Journal of Experimental Zoology 262:357–373.

    https://doi.org/10.1002/jez.1402620402

    • Google Scholar
14. 1. Esbjörnsson M
    2. Sylvén C
    3. Holm I
    4. Jansson E

    (1993) Fast twitch fibres may predict anaerobic performance in both females and males

    International Journal of Sports Medicine 14:257–263.

    https://doi.org/10.1055/s-2007-1021174

    • PubMed
    • Google Scholar
15. 1. Faircloth BC
    2. McCormack JE
    3. Crawford NG
    4. Harvey MG
    5. Brumfield RT
    6. Glenn TC

    (2012) Ultraconserved elements anchor thousands of genetic markers spanning multiple evolutionary timescales

    Systematic Biology 61:717–726.

    https://doi.org/10.1093/sysbio/sys004

    • PubMed
    • Google Scholar
16. 1. Friscia A
    2. Sanin GD
    3. Lindsay WR
    4. Day LB
    5. Schlinger BA
    6. Tan J
    7. Fuxjager MJ

    (2016) Adaptive evolution of a derived radius morphology in manakins (Aves, Pipridae) to support acrobatic display behavior

    Journal of Morphology 277:766–775.

    https://doi.org/10.1002/jmor.20534

    • PubMed
    • Google Scholar
17. 1. Fuxjager MJ
    2. Goller F
    3. Dirkse A
    4. Sanin GD
    5. Garcia S

    (2016a) Select forelimb muscles have evolved superfast contractile speed to support acrobatic social displays

    eLife 5:e13544.

    https://doi.org/10.7554/eLife.13544

    • PubMed
    • Google Scholar
18. 1. Fuxjager MJ
    2. Lee JH
    3. Chan TM
    4. Bahn JH
    5. Chew JG
    6. Xiao X
    7. Schlinger BA

    (2016b) Research resource: hormones, genes, and athleticism: effect of androgens on the avian muscular transcriptome

    Molecular Endocrinology 30:254–271.

    https://doi.org/10.1210/me.2015-1270

    • PubMed
    • Google Scholar
19. 1. Fuxjager MJ
    2. Schuppe ER
    3. Hoang J
    4. Chew J
    5. Shah M
    6. Schlinger BA

    (2016c) Expression of 5α- and 5β-reductase in spinal cord and muscle of birds with different courtship repertoires

    Frontiers in Zoology 13:25.

    https://doi.org/10.1186/s12983-016-0156-y

    • PubMed
    • Google Scholar
20. 1. Fuxjager MJ
    2. Fusani L
    3. Goller F
    4. Trost L
    5. Maat AT
    6. Gahr M
    7. Chiver I
    8. Ligon RM
    9. Chew J
    10. Schlinger BA

    (2017a) Neuromuscular mechanisms of an elaborate wing display in the golden-collared manakin (Manacus vitellinus)

    The Journal of Experimental Biology 220:4681–4688.

    https://doi.org/10.1242/jeb.167270

    • PubMed
    • Google Scholar
21. 1. Fuxjager MJ
    2. Miles MC
    3. Goller F
    4. Petersen J
    5. Yancey J

    (2017b) Androgens support male acrobatic courtship behavior by enhancing muscle speed and easing the severity of its tradeoff with force

    Endocrinology 158:4038–4046.

    https://doi.org/10.1210/en.2017-00599

    • PubMed
    • Google Scholar
22. 1. Garcia M
    2. Charrier I
    3. Rendall D
    4. Iwaniuk AN

    (2012) Temporal and spectral analyses reveal individual variation in a Non-Vocal acoustic display: the drumming display of the ruffed grouse (Bonasa umbellus, L.)

    Ethology 118:292–301.

    https://doi.org/10.1111/j.1439-0310.2011.02011.x

    • Google Scholar
23. 1. Goller F
    2. Suthers RA

    (1996) Role of syringeal muscles in controlling the phonology of bird song

    Journal of Neurophysiology 76:287–300.

    https://doi.org/10.1152/jn.1996.76.1.287

    • PubMed
    • Google Scholar
24. 1. Grant BR
    2. Grant PR

    (2010) Songs of Darwin's finches diverge when a new species enters the community

    PNAS 107:20156–20163.

    https://doi.org/10.1073/pnas.1015115107

    • Google Scholar
25. 1. Kettlewell HBD

    (1956) A resume of investigations on the evolution of melanism in the lepidoptera

    Proceedings of the Royal Society B: Biological Sciences 145:297–303.

    https://doi.org/10.1098/rspb.1956.0037

    • Google Scholar
26. 1. Koenker R
    2. Machado JAF

    (1999) Goodness of fit and related inference processes for quantile regression

    Journal of the American Statistical Association 94:1296–1310.

    https://doi.org/10.1080/01621459.1999.10473882

    • Google Scholar
27. 1. Komi P

    (1984) Physiological and biomechanical correlates of muscle function

    Exercise and Sport Sciences Reviews 12:81–122.

    https://doi.org/10.1249/00003677-198401000-00006

    • Google Scholar
28. 1. Kopp M
    2. Servedio MR
    3. Mendelson TC
    4. Safran RJ
    5. Rodríguez RL
    6. Hauber ME
    7. Scordato EC
    8. Symes LB
    9. Balakrishnan CN
    10. Zonana DM
    11. van Doorn GS

    (2018) Mechanisms of assortative mating in speciation with gene flow: connecting theory and empirical research

    The American Naturalist 191:1–20.

    https://doi.org/10.1086/694889

    • Google Scholar
29. 1. Lapiedra O
    2. Schoener TW
    3. Leal M
    4. Losos JB
    5. Kolbe JJ

    (2018) Predator-driven natural selection on risk-taking behavior in anole lizards

    Science 360:1017–1020.

    https://doi.org/10.1126/science.aap9289

    • PubMed
    • Google Scholar
30. 1. Lemey P
    2. Rambaut A
    3. Welch JJ
    4. Suchard MA

    (2010) Phylogeography takes a relaxed random walk in continuous space and time

    Molecular Biology and Evolution 27:1877–1885.

    https://doi.org/10.1093/molbev/msq067

    • PubMed
    • Google Scholar
31. 1. Lemmon AR
    2. Lemmon EM

    (2008) A likelihood framework for estimating phylogeographic history on a continuous landscape

    Systematic Biology 57:544–561.

    https://doi.org/10.1080/10635150802304761

    • PubMed
    • Google Scholar
32. 1. McDonald DB

    (1989) Correlates of male mating success in a lekking bird with male-male cooperation

    Animal Behaviour 37:1007–1022.

    https://doi.org/10.1016/0003-3472(89)90145-0

    • Google Scholar
33. 1. McDonald DB
    2. Clay RP
    3. Brumfield RT
    4. Braun MJ

    (2001) Sexual selection on plumage and behavior in an avian hybrid zone: experimental tests of male-male interactions

    Evolution 55:1443–1451.

    https://doi.org/10.1111/j.0014-3820.2001.tb00664.x

    • PubMed
    • Google Scholar
34. 1. Miles MC
    2. Fuxjager MJ

    (2018) Animal choreography of song and dance: a case study in the montezuma oropendola, Psarocolius montezuma

    Animal Behaviour 140:99–107.

    https://doi.org/10.1016/j.anbehav.2018.04.006

    • Google Scholar
35. 1. Miles MC
    2. Schuppe ER
    3. Ligon RM
    4. Fuxjager MJ

    (2018) Macroevolutionary patterning of woodpecker drums reveals how sexual selection elaborates signals under constraint

    Proceedings of the Royal Society B: Biological Sciences 285:20172628.

    https://doi.org/10.1098/rspb.2017.2628

    • Google Scholar
36. 1. Mowles SL
    2. Ord TJ

    (2012) Repetitive signals and mate choice: insights from contest theory

    Animal Behaviour 84:295–304.

    https://doi.org/10.1016/j.anbehav.2012.05.015

    • Google Scholar
37. 1. O'Donovan C
    2. Meade A
    3. Venditti C

    (2018) Dinosaurs reveal the geographical signature of an evolutionary radiation

    Nature Ecology & Evolution 2:452–458.

    https://doi.org/10.1038/s41559-017-0454-6

    • PubMed
    • Google Scholar
38. 1. Ord TJ
    2. Peters RA
    3. Clucas B
    4. Stamps JA

    (2007) Lizards speed up visual displays in noisy motion habitats

    Proceedings of the Royal Society B: Biological Sciences 274:1057–1062.

    https://doi.org/10.1098/rspb.2006.0263

    • PubMed
    • Google Scholar
39. 1. Pagel M
    2. Meade A
    3. Barker D

    (2004) Bayesian estimation of ancestral character states on phylogenies

    Systematic Biology 53:673–684.

    https://doi.org/10.1080/10635150490522232

    • PubMed
    • Google Scholar
40. Software

    1. Pagel M
    2. Meade A

    (2007) BayesTraits

    BayesTraits.

    http://www.evolution.rdg.ac.uk/BayesTraits.html
41. 1. Parchman TL
    2. Gompert Z
    3. Braun MJ
    4. Brumfield RT
    5. McDonald DB
    6. Uy JA
    7. Zhang G
    8. Jarvis ED
    9. Schlinger BA
    10. Buerkle CA
    11. Jac U

    (2013) The genomic consequences of adaptive divergence and reproductive isolation between species of manakins

    Molecular Ecology 22:3304–3317.

    https://doi.org/10.1111/mec.12201

    • PubMed
    • Google Scholar
42. 1. Parsons TJ
    2. Olson SL
    3. Braun MJ

    (1993) Unidirectional spread of secondary sexual plumage traits across an avian hybrid zone

    Science 260:1643–1646.

    https://doi.org/10.1126/science.260.5114.1643

    • PubMed
    • Google Scholar
43. 1. Pigot AL
    2. Tobias JA

    (2013) Species interactions constrain geographic range expansion over evolutionary time

    Ecology Letters 16:330–338.

    https://doi.org/10.1111/ele.12043

    • PubMed
    • Google Scholar
44. 1. Podos J

    (2001) Correlated evolution of morphology and vocal signal structure in Darwin's finches

    Nature 409:185–188.

    https://doi.org/10.1038/35051570

    • PubMed
    • Google Scholar
45. 1. Price JJ
    2. Earnshaw SM
    3. Webster MS

    (2006) Montezuma oropendolas modify a component of song constrained by body size during vocal contests

    Animal Behaviour 71:799–807.

    https://doi.org/10.1016/j.anbehav.2005.05.025

    • Google Scholar
46. Software

    1. Rambaut A
    2. Charleston M

    (2002) TreeEdit v1.0 Alpha 10

    TreeEdit v1.0 Alpha 10.

    http://tree.bio.ed.ac.uk/software/treeedit/main
47. 1. Reidy SP
    2. Kerr SR
    3. Nelson JA

    (2000)

    Aerobic and anaerobic swimming performance of individual Atlantic cod

    The Journal of Experimental Biology 203:347–357.

    • PubMed
    • Google Scholar
48. Software

    1. Ridgeley RS
    2. Allnutt TF
    3. Brooks T
    4. NcNicol DK
    5. Mehlman DW
    6. Young BE

    (2012)

    Digit Distrib Maps Birds West Hemisphere, version 5

    Digit Distrib Maps Birds West Hemisphere.
49. 1. Rivero JL
    2. Serrano AL
    3. Henckel P
    4. Agüera E

    (1993) Muscle fiber type composition and fiber size in successfully and unsuccessfully endurance-raced horses

    Journal of Applied Physiology 75:1758–1766.

    https://doi.org/10.1152/jappl.1993.75.4.1758

    • PubMed
    • Google Scholar
50. 1. Roff DA
    2. Fairbairn DJ

    (2007) The evolution of trade-offs: where are we?

    Journal of Evolutionary Biology 20:433–447.

    https://doi.org/10.1111/j.1420-9101.2006.01255.x

    • PubMed
    • Google Scholar
51. 1. Rome LC
    2. Syme DA
    3. Hollingworth S
    4. Lindstedt SL
    5. Baylor SM

    (1996) The whistle and the rattle: the design of sound producing muscles

    PNAS 93:8095–8100.

    https://doi.org/10.1073/pnas.93.15.8095

    • PubMed
    • Google Scholar
52. 1. Rudh A
    2. Rogell B
    3. Håstad O
    4. Qvarnström A

    (2011) Rapid population divergence linked with co-variation between coloration and sexual display in strawberry poison frogs

    Evolution 65:1271–1282.

    https://doi.org/10.1111/j.1558-5646.2010.01210.x

    • PubMed
    • Google Scholar
53. 1. Sanderson MJ

    (1997) A nonparametric approach to estimating divergence times in the absence of rate constancy

    Molecular Biology and Evolution 14:1218–1231.

    https://doi.org/10.1093/oxfordjournals.molbev.a025731

    • Google Scholar
54. 1. Schuppe ER
    2. Fuxjager MJ

    (2018) High-speed displays encoding motor skill trigger elevated territorial aggression in downy woodpeckers

    Functional Ecology 32:450–460.

    https://doi.org/10.1111/1365-2435.13010

    • Google Scholar
55. 1. Seddon N

    (2005) Ecological adaptation and species recognition drives vocal evolution in neotropical suboscine birds

    Evolution 59:200–215.

    https://doi.org/10.1111/j.0014-3820.2005.tb00906.x

    • PubMed
    • Google Scholar
56. 1. Servedio MR

    (2004) The evolution of premating isolation: local adaptation and natural and sexual selection against hybrids

    Evolution 58:913–924.

    https://doi.org/10.1111/j.0014-3820.2004.tb00425.x

    • PubMed
    • Google Scholar
57. 1. Servedio MR
    2. Burger R

    (2014) The counterintuitive role of sexual selection in species maintenance and speciation

    PNAS 111:8113–8118.

    https://doi.org/10.1073/pnas.1316484111

    • Google Scholar
58. 1. Servedio MR

    (2016) Geography, assortative mating, and the effects of sexual selection on speciation with gene flow

    Evolutionary Applications 9:91–102.

    https://doi.org/10.1111/eva.12296

    • PubMed
    • Google Scholar
59. 1. Sorci G
    2. Swallow JG
    3. Garland T
    4. Clobert J

    (1995) Quantitative genetics of locomotor speed and endurance in the lizard Lacerta vivipara

    Physiological Zoology 68:698–720.

    https://doi.org/10.1086/physzool.68.4.30166352

    • Google Scholar
60. 1. Syme DA
    2. Josephson RK

    (2002) How to build fast muscles: synchronous and asynchronous designs

    Integrative and Comparative Biology 42:762–770.

    https://doi.org/10.1093/icb/42.4.762

    • PubMed
    • Google Scholar
61. 1. Tello JG
    2. Moyle RG
    3. Marchese DJ
    4. Cracraft J

    (2009) Phylogeny and phylogenetic classification of the tyrant flycatchers, cotingas, manakins, and their allies (Aves: Tyrannides)

    Cladistics 25:429–467.

    https://doi.org/10.1111/j.1096-0031.2009.00254.x

    • Google Scholar
62. 1. Tobias JA
    2. Cornwallis CK
    3. Derryberry EP
    4. Claramunt S
    5. Brumfield RT
    6. Seddon N

    (2014) Species coexistence and the dynamics of phenotypic evolution in adaptive radiation

    Nature 506:359–363.

    https://doi.org/10.1038/nature12874

    • PubMed
    • Google Scholar
63. 1. Uy JA
    2. Borgia G

    (2000) Sexual selection drives rapid divergence in bowerbird display traits

    Evolution; International Journal of Organic Evolution 54:273–278.

    https://doi.org/10.1111/j.0014-3820.2001.tb01322.x

    • PubMed
    • Google Scholar
64. 1. Van Damme R
    2. Wilson RS
    3. Vanhooydonck B
    4. Aerts P

    (2002) Performance constraints in decathletes

    Nature 415:755–756.

    https://doi.org/10.1038/415755b

    • PubMed
    • Google Scholar
65. 1. Vanhooydonck B
    2. Van Damme R
    3. Aerts P

    (2001) Speed and stamina trade-off in lacertid lizards

    Evolution 55:1040–1048.

    https://doi.org/10.1554/0014-3820(2001)055[1040:SASTOI]2.0.CO;2

    • PubMed
    • Google Scholar
66. 1. Vanhooydonck B
    2. James RS
    3. Tallis J
    4. Aerts P
    5. Tadic Z
    6. Tolley KA
    7. Measey GJ
    8. Herrel A

    (2014) Is the whole more than the sum of its parts? Evolutionary trade-offs between burst and sustained locomotion in lacertid lizards

    Proceedings of the Royal Society B: Biological Sciences 281:20132677.

    https://doi.org/10.1098/rspb.2013.2677

    • Google Scholar
67. 1. Walker RS
    2. Ribeiro LA

    (2011) Bayesian phylogeography of the Arawak expansion in lowland South America

    Proceedings of the Royal Society B: Biological Sciences 278:2562–2567.

    https://doi.org/10.1098/rspb.2010.2579

    • Google Scholar
68. 1. Wilkins MR
    2. Seddon N
    3. Safran RJ

    (2013) Evolutionary divergence in acoustic signals: causes and consequences

    Trends in Ecology & Evolution 28:156–166.

    https://doi.org/10.1016/j.tree.2012.10.002

    • PubMed
    • Google Scholar
69. 1. Wilson RS
    2. James RS
    3. Van Damme R

    (2002)

    Trade-offs between speed and endurance in the frog xenopus laevis: a multi-level approach

    The Journal of Experimental Biology 205:1145–1152.

    • PubMed
    • Google Scholar
70. 1. Wilson DR
    2. Bitton PP
    3. Podos J
    4. Mennill DJ

    (2014) Uneven sampling and the analysis of vocal performance constraints

    The American Naturalist 183:214–228.

    https://doi.org/10.1086/674379

    • PubMed
    • Google Scholar

Author details

1. Meredith C Miles

   Wake Forest University, Winston-Salem, United States

   Contribution

   Formal analysis, Investigation, Methodology, Writing—original draft, Writing—review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-7307-0195

2. Franz Goller

   1. University of Utah, Salt Lake City, United States
   2. Institute for Zoophysiology, University of Münster, Germany

   Contribution

   Formal analysis, Investigation, Methodology, Writing—review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0001-5333-1987

3. Matthew J Fuxjager

   Wake Forest University, Winston-Salem, United States

   Contribution

   Conceptualization, Resources, Formal analysis, Supervision, Funding acquisition, Investigation, Methodology, Writing—original draft, Project administration, Writing—review and editing

   For correspondence

   mfoxhunter@gmail.com

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0003-0591-6854

• 2,858

  views

• 268

  downloads

• 27

  citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.