RESEARCH ARTICLE

The Effects of Slope and Branch Structure on the Locomotion of a Specialized Arboreal Colubrid Snake (Boiga irregularis) BRUCE C. JAYNE1* AND GREG BYRNES2 1

Department of Biological Sciences, University of Cincinnati, Cincinnati, Ohio Department of Biology, Siena College, Loudonville, New York

2

ABSTRACT

J. Exp. Zool. 323A:309–321, 2015

The surfaces in arboreal habitats have variable diameters, slopes, and branching structure that pose functional challenges for animal locomotion. Nevertheless, many lineages of snakes have independently evolved arboreality. We tested the effects of arboreal habitat structure on the locomotion of a highly arboreal species, the brown tree snake (Boiga irregularis), moving on seven diameters (0.6–21 cm) of cylinders oriented at three slopes (0°, 45°, 90°) and with or without pegs. Intermediate diameters of horizontal cylinders maximized speed, and some of the large-diameter cylinders without pegs were impassable when they were inclined. With increased slope the snakes were slower, and they changed from using lateral undulation with sliding contact and balancing to concertina locomotion with periodic static gripping. The presence of pegs increased the speeds of the brown tree snakes and resulted in them only using lateral undulation. Surface diameter, slope, and the occurrence of pegs also had widespread significant effects on the kinematics of the brown tree snakes. Overall, compared to anatomically less specialized corn snakes, brown tree snakes use more lateral undulation, are usually much faster, and are able to move on a wider variety of surfaces. Unlike some of the trade-offs found previously between two less specialized species of snakes with different stoutness when they used modes of arboreal locomotion that involved either balancing or gripping, the slender-bodied brown tree snakes excel at both. Hence, this species may not only be a “jack of all trades” but also a master of many. J. Exp. Zool. 323A:309–321, 2015. © 2015 Wiley Periodicals, Inc. How to cite this article: Jayne BC, Byrnes G. 2015. The effects of slope and branch structure on the locomotion of a specialized Arboreal Colubrid Snake (Boiga irregularis). J. Exp. Zool. 323A:309–321.

INTRODUCTION The surfaces in arboreal habitats impose many functional demands for the locomotion of animals beyond those commonly encountered in terrestrial habitats (Cartmill, '85). Nevertheless, a large number of phylogenetically diverse species move and live in trees, including animals with such profoundly different body plans as having or lacking limbs. Some of the difficulties of moving in arboreal habitats have similar expected effects for limbed and limbless animals. For example, moving up steep slopes increases the mechanical work that an animal must do against its own weight, although the magnitude of this effect depends on the size of the animal (Taylor et al., '72). When moving on top of narrow cylindrical surfaces, all animals without adhesive structures must either balance or grip to prevent toppling sideways (Cartmill, '85). By contrast, some other effects

of branch structure depend on the body plan of the animal. For example, secondary branches emerging from the primary surface create obstacles that impede the running of limbed animals

Grant sponsor: National Science Foundation; grant sponsor: IOS; grant number: 0843197.  Correspondence to: Bruce C. Jayne, Department of Biological Sciences, University of Cincinnati, PO Box 210006, Cincinnati 45221-0006, Ohio. E-mail: [email protected] Received 10 November 2014; Revised 19 December 2014; Accepted 20 January 2015 DOI: 10.1002/jez.1920 Published online XX Month Year in Wiley Online Library (wileyonlinelibrary.com).

© 2015 WILEY PERIODICALS, INC.

310 (Hyams et al., 2012; Jones and Jayne, 2012), whereas such structures help snakes generate propulsive forces with lateral undulation, during which weaving through branches can also reduce the tendency to topple (Astley and Jayne, 2009). In additional to different body plans among arboreal animals, quantitative variation within a body plan can modify the effects of habitat structure. For example, anole lizards with longer limbs prefer larger diameter branches and run faster than those species that have convergently evolved shorter limbs and a preference for narrower branches (Irschick and Losos, '99). However, shorter limbed species may slip less, which suggests that trade-offs occur between speed and slipping (Losos and Sinervo, '89). Highly arboreal species of snakes have also convergently evolved several quantitative anatomical traits including a light-weight body that seems well suited for enhancing the ability to crawl on slender branches and up steep slopes (Lillywhite and Henderson, '93; Pizzatto et al., 2007; Feldman and Meiri, 2013). Most of the previous comparative information on the arboreal locomotion of snakes is for corn snakes and boa constrictors (Astley and Jayne, 2009; Jayne and Herrmann, 2011), both of which move readily in trees. However, these species are much stouter than arboreal specialists such as the brown tree snake (Hoefer and Jayne, 2013), which is the focus of our study. Brown tree snakes also have long tails and axial muscles with long tendons, both of which have evolved convergently in a wide variety of arboreal species of colubroid snakes (Jayne, '82; Lillywhite and Henderson, '93), and the tail of brown tree snakes is prehensile. In addition to the considerable interspecific anatomical variation that is relevant to their locomotion, variation in locomotor behavior and performance occurs both for an individual snake on different surfaces and among different snake species (Gray, '46; Gans, '62; Jayne, '86). For example, during lateral undulation on solid surfaces, snakes propagate lateral bends posteriorly as the entire body of the snake slides over the supporting surfaces. By contrast, snakes periodically stop and grip during arboreal concertina locomotion (Byrnes and Jayne, 2014). For the arboreal locomotion of juvenile snakes, corn snakes are faster than boa constrictors on cylindrical surfaces with pegs when both species use lateral undulation, but boa constrictors are faster on inclined cylinders without pegs where both species perform concertina locomotion (Jayne and Herrmann, 2011). Boa constrictors are much heavier and more muscular than corn snakes (Hoefer and Jayne, 2013), which suggests that anatomical trade-offs exist for the abilities to grip strongly and move rapidly with lateral undulation. On horizontal cylinders lacking pegs, corn snakes use concertina locomotion, whereas boa constrictors use lateral undulation on narrow cylinders and concertina on wide cylinders (Astley and Jayne, 2007; Jayne and Herrmann, 2011). For the one diameter that has been investigated, brown tree snakes use lateral undulation on horizontal cylinders (Crotty and Jayne, 2015). J. Exp. Zool.

JAYNE AND BYRNES To test the effects of branch structure on the locomotion of brown tree snakes, we manipulated the diameter of cylindrical surfaces and used surfaces with varying slopes and with pegs to simulate secondary branches. We expected the brown tree snakes to use lateral undulation when pegs were present and concertina locomotion on steep cylinders lacking pegs, which would prevent slipping downward. Whether brown tree snakes would use lateral undulation across a wide range of diameters of horizontal cylinders lacking pegs was much less clear. Based on previous studies, we expected the speeds of brown tree snakes to decrease with increased slope as well as with increasing diameters of cylinder greater than the diameter of the of the snake. We also expected the presence of pegs to enhance the speeds of the snakes and enable them to climb large diameter cylinders that would otherwise be impassable. As the structure of surfaces changed we expected the snakes to make postural adjustments that facilitate balancing, gripping the perch, or transmitting forces to the supporting surface. We also compared our results for brown tree snakes to those for corn snakes on identical surfaces (Astley and Jayne, 2009). Given that brown tree snakes are anatomically and behaviorally more specialized than corn snakes for arboreal habitats and that a lighter body may be beneficial for enhancing the speeds of lateral undulation on solid surfaces (Ruben, '77), we expected brown tree snakes to have faster arboreal lateral undulation. The trade-offs found previously for stouter boa constrictors that climb faster when gripping and using concertina locomotion than the more slender corn snakes (Jayne and Herrmann, 2011) suggest that brown tree snakes, which are even more slender and with longer axial muscle segments than corn snakes (Hoefer and Jayne, 2013), may also have a trade-off that causes slower speeds of concertina locomotion compared to corn snakes. However, one caveat for this prediction is that prehensile tails are present in brown tree snakes but absent in corn snakes, and this structure seems well suited for enhancing the ability to grip during arboreal concertina locomotion.

MATERIALS AND METHODS Experimental Subjects All of the brown tree snakes were captured in Guam and transported to the laboratory at the University of Cincinnati where all of the experiments were conducted. We used 10 adult brown tree snakes (Boiga irregularis) with sizes that were as uniform as was practical to obtain: snout–vent length (mean SVL ¼ 109 cm, range 95–118 cm), total length (mean TL ¼ 139 cm, range 121–152 cm) and mass (mean ¼ 176 g, range 140–203 g). The sexes of the brown tree snakes were unknown because determining sex definitively in the rather large individuals of this species can require invasive methods that could injure the animals. To facilitate comparisons with previously studied corn snakes, Pantherophis guttatus (Astley

BRANCH STRUCTURE AFFECTS SNAKE LOCOMOTION

311

and Jayne, 2009), we also selected brown tree snakes so that their lengths would be similar to those of the corn snakes, for which mean SVL ¼ 113 cm (range 110–117 cm), mean TL ¼ 134 cm (range 132–140 cm) and mean mass ¼ 443 g (range 381–485 g). The captive snakes were housed individually in cages (26  59  57 cm) with incandescent light bulbs that allowed the snakes to regulate their body temperature from 25 to 33°C. Depending on the time of day and microhabitat, free ranging brown tree snakes in Guam commonly have body temperatures ranging from 21 to 31°C, and when sufficient heat sources are available this species commonly regulates body temperatures between 28 and 31°C (Anderson et al., 2005). Prior to testing, we marked all snakes with white paint at approximately 25, 50, 75, and 100% of SVL to provide landmarks for video analysis. The body temperatures of the snakes during the experiments were between 29 and 31°C. The tests of performance were conducted at least seven days after the most recent feeding of the snakes, which was sufficient time for them to complete digestion and defecation from the last meal (mouse mass ¼ 25–40 g). All procedures and animal care were in accordance with guidelines by the Institutional Animal Care and Use Committee of the University of Cincinnati (Protocol 07-01-08-01#). Experimental Apparatus and Procedures We used a total of 31 combinations of surfaces and slopes to determine the locomotor performance of the brown tree snakes (Table 1). The seven diameters of the primary cylindrical surface supporting the snakes ranged from 0.6 to 21.0 cm, and the long axis of each cylinder was oriented horizontally (0°) or inclined 45° or 90° relative to horizontal. When pegs were present, they were placed at 10 cm intervals along the top center line of the primary cylindrical surface, and the long axis of each peg was perpendicular to that of the primary surface. Each peg had a diameter of 6.5 mm and protruded 38 mm beyond the surface of

Table 1. Summary of the surfaces used for the brown tree snakes in this study, and those in common with a previous study of corn snakes (Astley and Jayne, 2009) are indicated in bold italicized type. Slope Diameter (cm) 0.6 1.6 2.9 4.1 8.9 15.9 21.0



45°

90°

P, þP P, +P P P, +P P P, +P P

P, þP P, +P P P, þP P P, þP

P, þP P, +P P P, þP P P, þP

þP, pegs present; P, pegs absent.

the primary cylinder. Lines drawn every 10 cm along the primary surface provided a distance scale. To permit direct comparisons between species, we covered all surfaces with the same (Nashua 394) duct tape (Franklin, KY, USA) as in previous laboratory studies of corn snakes (Astley and Jayne, 2007; Astley and Jayne, 2009). The fiber mesh in the tape created a uniform texture for all of the pipes with different diameter that were used and a roughness greater than that of the underlying materials. The roughness also appeared to be within the range of at least some types of tree bark with branches of a similar diameter. Furthermore, the coefficient of friction (mean ¼ 0.28, range 0.23–0.39) between this tape and snake skin (Astley and Jayne, 2007) resembles that for a wide range of materials and roughness (Gray and Lissmann, '50). After placing the snakes on the test apparatus, we lightly tapped their tails to encourage them to attain maximal speeds. For a given surface, we usually performed three trials in rapid succession (

The effects of slope and branch structure on the locomotion of a specialized arboreal colubrid snake (Boiga irregularis).

The surfaces in arboreal habitats have variable diameters, slopes, and branching structure that pose functional challenges for animal locomotion. Neve...
828KB Sizes 3 Downloads 7 Views