Pierre Legreneur, 1,2 * Dominique G. Homberger, 3 and Vincent Bels 1

JOURNAL OF MORPHOLOGY 273:765 775 (2012) Assessment of the Mass, Length, Center of Mass, and Principal Moment of Inertia of Body s in Adult Males of the Brown Anole (Anolis sagrei) and Green, or Carolina, Anole (Anolis carolinensis) Pierre Legreneur, 1,2 * Dominique G. Homberger, 3 and Vincent Bels 1 1 UMR 7179 du CNRS, Département Ecologie et Gestion de la Biodiversité, Muséum National d Histoire Naturelle, Paris, France 2 Centre de Recherche et d Innovation sur le Sport, Université de Lyon, Villeurbanne, France 3 Department of Biological Sciences, Louisiana State University, Baton Rouge, Louisiana 70803 ABSTRACT This study provides a morphometric data set of body segments that are biomechanically relevant for locomotion in two ecomorphs of adult male anoles, namely, the trunk-ground Anolis sagrei and the trunkcrown Anolis carolinensis. For each species, 10 segments were characterized, and for each segment,, mass, location of the center of mass, and radius of gyration were measured or calculated, respectively. The radii of gyration were computed from the moments of inertia by using the double swing pendulum method. The trunkground A. sagrei has relatively longer and stockier hindlimbs and forelimbs with smaller body than A. carolinensis. These differences between the two ecomorphs demonstrated a clear relationship between morphology and performance, particularly in the context of predator avoidance behavior, such as running or jumping in A. sagrei and crypsis in A. carolinensis. Our results provide new perspectives on the mechanism of adaptive radiation as the limbs of the two species appear to scale via linear factors and, therefore, may also provide explanations for the mechanism of evolutionary changes of structures within an ecological context. J. Morphol. 273:765 775, 2012. Ó 2012 Wiley Periodicals, Inc. KEY WORDS: squamates; ecomorph; locomotion, Anolis, jumping INTRODUCTION Anoles (Anolis spp.) represent one of the largest genera of lizard-like reptiles (Squamata) with over 400 species and occur on the American continents, including the Caribbean (Poe, 2004; Losos, 2009). This group is characterized by one of the greatest morphological, behavioral, and ecological diversity among arboreal tetrapods and has been used as a major model for analysing evolutionary processes in animal communities (Yang et al., 1974; Losos, 1994; 2009; Butler and King, 2004; Knouft et al., 2006). Anoles can be classified into various ecomorphological groups based on the part of their habitat they use predominantly, such as the height above ground or the diameter of perches. Accordingly, these ecomorphs are designated as grass-bush, trunk-ground, trunk, twig, trunk-crown, and crown-giant (Williams, 1972). These basic ecomorphs have been used to study the interactions between locomotor performance and ecological radiation in tetrapods (Pounds, 1988; Losos, 1990a; 2009; Irschick and Losos, 1998; Mattingly and Jayne, 2005). So far, various species of anoles have been used to analyze the functional and ecological significance of some of the morphological, locomotor, and ecological traits within a phylogenetic context (Toro et al., 2004; Toro et al., 2006; Vanhooydonck et al., 2006b). However, all these studies provide no quantitative information on the force production mechanisms by the musculo-skeletal systems in response to the physical constraints of the environment. Among anoles, the trunk-ground Brown Anole (Anolis sagrei) and the trunk-crown Green Anole (Anolis carolinensis) have been used to analyze the functional consequences of variation in the morphology of the limbs (e.g., limb ), locomotor capabilities (e.g., jumping and running), and characteristics of their environment (Bels et al., 1992; Spezzano and Jayne, 2004). The two species differ in the relative frequency of walking and jumping in their locomotion behavior (Losos, 2009) to catch Additional Supporting Information may be found in the online version of this article. Contract grant sponsor: Maëlle Deloche; Contract grant sponsor: Claude Meunier; Contract grant sponsor: Lionel Reveret and Paul- Antoine Libourel (ANR project Kameleon); Grant number: ANR-05- MMSA0002-02 Masse de Données. *Correspondence to: Pierre Legreneur, UMR 7179 CNRS-MNHN, 57 Rue Cuvier, Case postale 55 F 2 75231 Paris Cedex 5 France. E-mail: pierre.legreneur@univ-lyon1.fr Received 22 June 2011; Revised 30 January 2012; Accepted 18 February 2012 Published online 28 March 2012 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/jmor.20022 Ó 2012 WILEY PERIODICALS, INC.

766 P. LEGRENEUR ET AL. prey (Montuelle et al., 2008) and to escape predators (Mattingly and Jayne, 2005). The two species also differ in the maximum performance of jumps between perches despite their similar snout-vent s (SVL), as A. sagrei manages to jump farther than A. carolinensis (Toro et al., 2003). Because the difference in the total limb of the two species is minor, we hypothesized that a difference in the power muscles of their limbs may account for the difference in performance. To test this hypothesis, it is necessary to evaluate the dynamics of the propulsion phase of a jump in the two species. This can be achieved by recording the displacement of the major limb joints as well as the ground reaction force during propulsion of the body by the hind feet (Toro et al., 2006). But these data are not sufficient to compute the forces and torques that act at each joint of the hindlimb and, thus, the contributions of the agonist and antagonist actuators at each joint. To do so, an inverse dynamics procedure can help to analyze the dynamics of a jumping lizard based on data of the ground reaction force and the kinematics (e.g., linear and angular displacement, velocity, and acceleration) at each joint (Legnani et al., 1996a,b). This inverse dynamic procedure consists in computing the forces and torques acting on each body segment from the kinematics and dynamics of a jump, as well as the morphometric characteristics of each segment. It is, therefore, necessary to analyze the morphometry (i.e., the and mass of the various limb segments, the location of their center of mass, and the radius of gyration) of the animals whose jumping performance is studied. To analyze the relationships between morphological traits and their performances, we compared a suite of morphometric data of features. These features are functionally and biomechanically relevant for the locomotor performances of the two species, such as maintaining particular postures on the substratum, walking, running, or jumping. The goal of this study was to provide a basic model for the two species, through which the morphometric characteristics of all segments can be calculated based on the SVL and total body mass of individuals without the need to sacrifice them. For that reason, the data presented in this article will be normalized to these two basic characteristics. As a first step to creating such a model and to minimize the number of variables, our morphometric study included only male individuals. METHODS Specimens Specimens of two anole species were obtained from a commercial dealer (La Ferme Tropicale, Paris, France). Only male individuals were used in this study to control for the morphometric variability between the sexes (Irschick et al., 2005a). Five specimens each of Anolis sagrei and Anolis carolinensis were sacrificed by intraperitoneal injection of 0.1-ml pentobarbital and frozen with the agreement of the Committee of the Muséum National d Histoire Naturelle. To prevent the retraction of the muscles during the dissection, which would falsify the positions of the centers of masses and moments of inertia, each specimen was frozen. After 24 h of freezing, a decrease of the mass of individuals was observed due to dehydration of the body. These losses of mass were in the order of 11.5 6 9.9% in A. sagrei and 12.7 6 10.5% in A. carolinensis. On the assumption that dehydration was proportional in all segments of the body, no conversion factor was applied to the results presented here. The system of Linked s Modeling an animal body in n segments depends on the specific questions that one needs to solve regarding the moving organism. During locomotion, the movement and orientation in space of each segment depends on the particular movement and locomotor mode of the specimen. Therefore, the degree of detail of a model of each segment depends on the particular movement of the organism. For example, the analysis of the grasping of a branch by the forelimbs requires a detailed modeling of each finger, whereas the analysis of a jump powered by the hindlimbs allows a more general modeling of the fingers. The degree of detail needed to model the trunk is less obvious. At first glance, it seems necessary to consider the trunk as a construct of three segments, namely the thorax, the abdomen, and the pelvis. However, based on preliminary data we assume that trunk movements during jumping and walking result in negligible bending movements of the segment between the glenohumeral joint of the forelimb and the greater trochanter of the hindlimb are small. Therefore, we have defined the trunk as a single segment between the first cervical vertebra and the first caudal vertebra. This vertebra is the first vertebra of the tail with a chevron and is located just after the third pygal vertebra (Hoffstetter and Gasc, 1969; Hyman, 1992). Fourteen anatomical landmarks were used to define the system of linked segments (Fig. 1): 1) Snout cranial end, 2) first cervical vertebra, 3) glenohumeral joint, 4) greater trochanter, 5) vent, 6) first caudal vertebra, 7) tail caudal end, 8) elbow joint, 9) wrist joint, 10) distal ends of digits, 11) femoral condyle, 12) talo-crural joint, 13) metatarso-tarsal joint, and 14) distal end of second toe. These landmarks define 10 segments: 1) head, 2) trunk, 3) tail, 4) arm, 5) forearm, 6) hand, 7) thigh, 8) Leg, 9) tarsus, and 10) metatarsus (Table 1). Each segment was defined by its proximal and distal, or cranial and caudal, ends, respectively. The segments of the vertebral column forming the mid-sagittal axis (i.e., the trunk, head-trunk, headtrunk-tail) and the head-forearms-trunk-tail system were defined as being between the glenohumeral joint and the greater trochanter. These definitions allows the modeling of the trunk segments with the help of easily identifiable markers on the video frames without having to digitize the snout, the first cervical vertebra, the first caudal vertebra, or the caudal end of the tail. Preparation and Isolation of s All segments were isolated from one another at their joints by disarticulating the joints under a stereoscopic microscope (Leica EZ4, 83 magnification). After cutting the skin, muscles and ligaments with scissors, the joint was revealed, and the articular capsule could be cut so that the joint surfaces could simply be separated. Measurements and Calculations of the Mass and Length of s The mass of the segments was measured with a Mettler Toledo balance AE100 (accurate to 0.1 mg, weighing range 0 109 g). Because of the rapid dehydration of the segments, all

MORPHOMETRIC CHARACTERIZATION OF ANOLES 767 Fig. 1. Diagrammatic representation of the linkage system in anoles. 1: Snout cranial end; 2: First cervical vertebra; 3: Glenohumeral joint; 4: Greater trochanter; 5: Vent; 6: First caudal vertebra; 7: Tail caudal end; 8: Elbow joint; 9: Wrist joint; 10: Digit distal end; 11: Femoral condyle; 12: Talocrural joint; 13: Metatarsotarsial joint; and 14: Distal end of second toe. segments were weighed during the same session. The segment mass coefficient k mi of segment i was computed from the ratio of the segment mass m i and the lizard total body mass M: m i ¼ k mi M The of each segment was measured with Rocky digital callipers (accurate to 0.1 mm). The segment coefficient k li of segment i was computed from the ratio of the segment l i and the SVL of the lizard: l i ¼ k li SVL The of the tail was not considered for normalizing the data because of its extreme variability, e.g., due to tail regeneration after autotomy (Cox, 1969a,b). Determination of the Center of Mass The center of mass was determined by using a specific device for small animals (Wells and DeMenthon, 1987; Supporting Information). The data describing the center of mass, expressed in percent of the segment, correspond to the relative distance between the proximal or cranial end, and the center of mass of a segment. For the trunk segment, the cranial end was defined by the glenohumeral joint of the forelimb and the caudal end by the greater trochanter of the hindlimb. ð1þ ð2þ Although the position of the center of mass of each segment was located experimentally, the position of the center of mass of poly-articular chains (e.g., head-trunk-tail system, forelimb, etc.) was calculated geometrically from the center of mass of the constituting segments. In a poly-articular chain composed of n segments and n 1 1 joints (i [ {0...n}), A 0 is the most proximal or cranial end of the chain (Fig. 2). If k i represents the center of mass G i of a segment i relative to the of the segment, the center of mass G of the whole system relative to A o is given by: ƒƒ! 1 X n A 0 G ¼ P n i¼1 m m i A ƒ ƒ! P n 0 G i ¼ i¼1 kmia ƒ ƒ! P 0 G i n i i¼1 i¼1 k mi The center of mass G i of the segment i relative to A o is given by ƒ ƒ! ƒƒƒƒ! X i A 0 G i ¼ ð ki 1ÞA i 1 A i þ j¼1 ƒƒƒƒ! A j 1 A j ƒƒ! Introducing (2) ƒƒƒ! and (4) into (3) allows the expression of A 0 G as a function of A 0 A n: ƒƒ! ƒƒƒ! A 0 G ¼ KA0 A n ð5þ with ð3þ ð4þ TABLE 1. Definition of the segments by their proximal and distal, or cranial and caudal ends, respectively Proximal end Distal end Hindlimb Greater trochanter Distal end of second toe Thigh Greater trochanter Femoral condyles Leg, tarsus, and metatarsus Femoral condyles Distal end of second toe Leg Femoral condyles Talo-crural joint Tarsus and metatarsus Talo-crural joint Distal end of second toe Tarsus Talo-crural joint Metatarso-tarsal joint Metatarsus Metatarso-tarsal joint Distal end of second toe Forelimb Glenohumeral joint Digit distal end Arm Glenohumeral joint Elbow joint Forearm and hand Elbow joint Digit distal end Forearm Elbow joint Wrist joint Hand Wrist joint Digit distal end Head Snout cranial end First cervical vertebra Trunk Glenohumeral joint Greater trochanter Tail First caudal vertebra Tail caudal end Head and trunk Glenohumeral joint Greater trochanter Head, trunk, and tail Glenohumeral joint Greater trochanter Head, forelimbs, trunk, and tail Glenohumeral joint Greater trochanter

768 P. LEGRENEUR ET AL. Fig. 2. Diagrammatic representation of the poly-articular chain used for calculating its center of mass from the positions of the individual center of mass Gi of the segment Ai-1Ai, i [ {1...n}<i/i>. P h n i¼1 k mi ðk i 1Þk li þ P i K ¼ P n i¼1 k P n li i¼1 k mi j¼1 k lj i ð6þ their proximal and distal ends. The relationship between these moments of inertia and those about the center of mass is given by the parallel-axis theorem (Winter, 2009): Estimation of the Radius of Gyration The principal moment of inertia of each segment along its longitudinal axis relative to its center of mass was identified with a double swing pendulum apparatus (Supporting Information), which was composed of: A tray (mass m 5 0.637 g, moment of inertia j, 2R 5 31 mm, width 5 19.7 mm), which is suspended by nylon wires. Two parallel suspension nylon wires ( L 5 440 mm, diameter 5 0.12 mm). Their mass was negligible compared to the mass of the tray. A segment on the tray (mass m i, moment of inertia J i ). The segment was placed orthogonally to the of the tray, and the center of mass of the segment was aligned with the center of mass of the tray. The guiding principle was to estimate the average period of five twisting oscillation of the tray without and with a segment. To do this, the angular displacement of the tray was recorded with a high-frequency video camera (Sanyo VPC-HD2000A, 240Hz) and tracked with a motion analyser created with Matlab 1 7.3.0 software (MathWorks, Natick, MA). The procedure was repeated five times for each segment. The mean period of oscillations was calculated for all motions of each segment. The equation for the period T was given as follows (Wells and DeMenthon, 1987): T 2 ¼ 4p 2 ðj i þ jþl ðm i þ mþgr 2 ð7þ Thus, the subtraction of the mean oscillation periods of the tray (calculation of j) from those of the tray-cum-segment (calculation of J i 1 j) allows the determination of the moment of inertia of a segment by itself (J i ). From the moment of inertia of a segment, it is possible to calculate the radius of gyration r i of a segment at its center of mass: r i ¼ sffiffiffiffiffiffi J i m i The segments of the hand, forearm, tarsus, and toes were modeled as cylinders because of their extremely low weights. Thus, their moments of inertia are given by: ð8þ J i ¼ 1 12 m il 2 i ð9þ These measurements provide the moment of inertia of a segment only about its center of mass. Because in reality most segments do not rotate about their center of mass, but rather about their joints, we calculated the moment of inertia about J ¼ J 0 þ mx 2 ð10þ where J is the moment of inertia about the proximal or distal end of a segment, J o is the moment of inertia about the center of mass, m is the mass of the segment, and x is the distance between the center of mass and the proximal or distal end. The radius of gyration is presented as a function of the of a segment. For the segments along the vertebral mid-sagittal axis (trunk, head-trunk, head-trunk-tail) and the head-forearms-trunk-tail system, the was defined as the distance between the glenohumeral joint of the forelimb and the greater trochanter of the hindlimb. Statistical Analysis The statistical analysis was performed on the raw data for the SVL and total body mass. For all other parameters, the tests were conducted on the relative values of mass,, center of mass, and radius of gyration of the segments. To compare the differences between the two ecomorphs, the nonparametric Kruskall-Wallis one-way analysis of variance was used. Statistical significance was accepted and considered as significant if the probability was less than 0.05 for each test (P < 0.05). RESULTS The specimens of A. carolinensis in our sample were significantly longer than the specimens of A. sagrei (SVL: 57.4 6 3.9 mm vs. 51.5 6 3.3 mm, P < 0.05). However, the two species had similar weights (4.30 6 1.11 g vs. 3.55 6 0.62 g, p 5 0.347). All morphometric data of anoles are given in Tables A1 A3 in the appendix. A. sagrei had relatively longer and stockier hindlimbs than A. carolinensis, and the individual segments of the hind limbs were also longer (Fig. 3) and heavier (Fig. 4) in A. sagrei. Among the measurements (relative to body mass and SVL), only those of the and mass of the metatarsus and the mass of the tarsus were similar in the two species. In addition, the relative positions of the centre of mass (Fig. 5) and radius of gyration (Fig. 6) of the hindlimbs were also similar in the two species. The forelimbs, like the hindlimbs, are longer (Fig. 3) and heavier (Fig. 4) in A. sagrei than in A. carolinensis. However, except for the of the forearm, the mass and of the individual

MORPHOMETRIC CHARACTERIZATION OF ANOLES 769 The and mass of the individual segments relative to their corresponding body parts were similar in A. sagrei and A. carolinensis (Fig. 7). Thus, in both species, the head, trunk, and tail represented 13, 30, and 58%, respectively, of the of the head-trunk-tail system and 20, 68, and 12%, respectively, of the mass of this system. The arm, forearm, and hand represented 36, 29, and 35%, respectively, of the of the forelimb and 53, 31, and 16, respectively of the mass of the forelimb. Similarly, the thigh, leg, and tarsusmetatarsus represented 28, 29, and 42%, respectively, of the of the hindlimb and 57, 29, and 14%, respectively, of the mass of the hindlimb. In contrast, the ratios of the of the tarsus and metatarsus were not the same in the two species. The tarsus was relatively longer and the metatarsus relatively shorter in A. sagrei than in Fig. 3. Comparison of segment coefficients of the hindlimb, forelimb, and head-trunk-tail in A. sagrei (black) and A. carolinensis (gray). Stars indicate a significant difference between the two species (P < 0.05). segments of the fore limbs were similar in the two species. The relative locations of centre of mass (Fig. 5) and radius of gyration (Fig. 6) of the forelimbs were also similar in the two species. The segments of the vertebral mid-sagittal axis, however, differed in the two species. A. sagrei had a shorter head and longer trunk than A. carolinensis (Fig. 3). The mass of the head-forelimbs-trunktail and head-trunk-tail systems was less in A. sagrei than in A. carolinensis (Fig. 4). These two segments were the only segments that differed significantly in the location of their center of mass and radius of gyration in the two species. The center of mass of the trunk in A. sagrei was located significantly more caudally and closer to the greater trochanter joint than in A. carolinensis (Figs. 5 and 7). Moreover, the radius of gyration of the head and trunk was significantly longer in A. sagrei (Fig. 6). Fig. 4. Comparison of segment mass coefficients of the hindlimb, the forelimb and the head-trunk-tail system in A. sagrei (black) and A. carolinensis (gray). Stars indicate a significant difference between the two species (P < 0.05).

770 P. LEGRENEUR ET AL. et al., 1992; Toro et al., 2003; Irschick et al., 2005b; Vanhooydonck et al., 2005) and A. sagrei (Losos et al., 2000; Spezzano and Jayne, 2004; Herrel et al., 2008). However, unlike most studies that present morphometric data of anoles, this study calculated the relative values of all segments of the body, because it is easier to interpret animal motor performance in terms of relative morphometric characteristics rather than in terms of absolute ones (Scholz et al., 2006). For example, during a jump, which is one of the main motor behaviors to escape predators and to catch prey (Montuelle et al., 2008), muscular work equivalent to the mechanical work necessary to displace the center of mass vertically during the take-off and airborne phases needs to be generated. In arboreal amniote tetrapods, only the hindlimbs participate in the acceleration of the body s center of mass during the last 70% of the take-off phase (Legre- Fig. 5. Comparison of the location of the center of mass coefficients in A. sagrei (black) and A. carolinensis (gray). These coefficients represent the relative distance between the proximal end of a segment and its center of mass. Stars indicate a significant difference between the two species (P < 0.05). A. carolinensis. Hence, the metatarsus was relatively heavier in A. carolinensis than in A. sagrei. DISCUSSION This study compared the morphometric characteristics of two species of anoles that represent different ecomorphs: The trunk-crown Green Anole (A. carolinensis) and the trunk-ground Brown Anole (A. sagrei). The main goal of our study was to collect relative measurements of body segments to create general, idealized models of these species, on the basis of which the dynamics of their locomotion could be analyzed through an inverse dynamic procedure. In absolute values, the body mass and SVL of the studied specimens were similar to those reported in the literature for A. carolinensis (Bels Fig. 6. Comparison of radius of gyration coefficients about the center of mass of the segments in A. sagrei (black) and A. carolinensis (gray). Stars indicate a significant difference between the two species (P < 0.05).

MORPHOMETRIC CHARACTERIZATION OF ANOLES 771 Fig. 7. Comparison of the body shapes of two ecomorphs, the Brown Anole (A. sagrei; brown area in A and brown outline in B) and the Green Anole (A. carolinensis; green area in A and green outline in B). The black dot represents the location of the vent. The of the segments are expressed in function of the SVL for each specie. The SVL are consequently identical (A) Realistic representation of the shape of the two ecomorphs. The yellow-black dots represent the positions of the centers of mass of the individual segments. (B) Diagrammatic representation of the two species. The of the segments is drawn to scale, but the width of the segments is drawn proportional to their mass to show the difference in the muscle volume of the forelimbs and hindlimbs of the two species. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.] neur et al., 2012). Therefore, the capacity of the hindlimbs to generate mechanical power is critical for jumping. Power generation results from a trade-off between different parts of the musculoskeletal system of the limbs (Toro et al., 2004; James et al., 2007). The longer the limb segments are (Harris and Steudel, 2002), the longer are the lever arms for the muscles across the joints (Jacobs and van Ingen Schenau, 1992) and, as long as the tendon-to-muscle ratio is positive for the extensor muscles (Biewener and Roberts, 2000), the more efficient is the power generation. However, the intra- and interspecific allometric analysis of hindlimb power production has to take into account the total body mass of an organism. For example, if two species of different body weights and sizes presented similar hindlimb morphometric characteristics, it could be expected that their jumping abilities would differ. Therefore, absolute morphometric values need to be scaled to the weights and total body s of the studied specimens. The major differences between the two studied ecomorphs were observed in the mass and of the segments of the hindlimbs. These segments are relatively longer and heavier in A. sagrei than in A. carolinensis. Assuming that the contractile properties of extensor muscles of the hindlimbs are similar, it would follow that A. sagrei would be able to develop a greater mechanical power per body mass unit than A. carolinensis, especially because of their similar body weights in this study. However, previous study on power generation during jumping in anoles showed that A. carolinensis was able to produce greater amount of specific muscle power than A. sagrei (Vanhooydonck et al., 2006a). That clearly demonstrates that morphometrics alone are not enough to predict muscle power output, and thus whole animal performance. Other parameters such as muscle biomechanics (typology, architecture) and the terms of muscle activation may also differ between the two species. Despite these morphometric differences, the relative positions of the center of mass and radius of gyration of the hindlimb segments are similar in the two species. This suggests that the distribution of the hindlimb mass is similar in the two species, i.e., half the mass of the hindlimb is spread over the first one-third of its at the knee. In the forelimbs, significant differences between the two species were observed only in the mass and of the whole limbs. As generalized leapers, anoles use their forelimbs during the take-off phase to orient the velocity vector of their center of mass at the beginning of the displacement (Legreneur et al., 2012) and to land (Bels et al., 1992; Crompton and Sellers, 2007; Legreneur and Bels, 2010). Since A. sagrei is probably capable of achieving a greater leap performance (Toro et al., 2003) and has a greater body mass than A. carolinensis, its longer forelimbs can absorb the greater impact force at the instant of landing (Losos, 1990b). As in the hindlimb, the locations of the center of mass of the segments of the forelimb and their radii of rotation indicate that the mass distribution is similar along the main axis of the segments in the two species. In the head-trunk-tail system, the trunk is relatively longer in A. sagrei, and the head is relatively longer in A. carolinensis. Nevertheless, the masses of the head and trunk (relative to the total body mass) are the same in the two species, even though the distribution of mass differs within these segments as is shown by the different positions of the center of mass and the of the radii of gyration in the two species (i.e., the center of mass of the trunk is located more caudally, and the radii of gyration are longer for the head and trunk, in A. sagrei than in A. carolinensis). Thus, the two anole

772 P. LEGRENEUR ET AL. ecomorphs clearly differ from each in their morphology (Losos, 1990a; Irschick et al., 1997) (Fig. 7). To summarize, A. sagrei has a smaller body with longer and thicker limbs than A. carolinensis. When the form and function of organisms are integrated and compared between species, ecological and evolutionary aspects need to be considered (Bock, 1994). For example, A. sagrei and A. carolinensis are classified as trunk-ground and trunk-crown ecomorphs, respectively (Losos, 2009). A. sagrei moves between two substrates with different physical and spatial constraints, namely the horizontal ground and vertical tree trunks, and moves usually on branches and trunks of large diameters (Losos et al., 2000). To escape predators, it is more effective to jump from the ground directly to the trunk than to run first toward the tree and then climb up the trunk, even though it is energetically more expensive. A. sagrei, whose limbs are relatively long as shown in this study, does indeed tend to use locomotor strategies that require relative long hindlimbs and forelimbs (Moermond, 1979), such as jumping. Because the maximal shortening velocity of a muscle is mechanically directly related to its, A. sagrei also relies on the maximal angular velocity of its joints and, thus, on maximal speed to escape predators (Irschick and Losos, 1998). In contrast, adult A. carolinensis lives mainly on broad surfaces of tree branches and uses various strategies to avoid detection by predators, such as changing its color to match the background, flattening its body against the surface, sidling to the other side of a branch or trunk, running, or jumping (Irschick and Losos, 1998). Expressed in terms of SVL or total body mass, the morphometric characteristics of A. carolinensis and A. sagrei are related to their respective behaviors and modes of locomotion in adaptation to the structure of their habitats. Although the two ecomorphs are similar in terms of the relative and mass of most of their segments, they differ significantly in the relative of their tarsus and metatarsus, with A. sagrei having a longer tarsus and shorter metatarsus than A. carolinensis. These differences could be related to the differences in the diameter of their preferred substrate, with A. sagrei moving on supports of larger diameters than A. carolinensis (Losos et al., 2000). However, in both species, 57% of the mass of the hindlimbs was concentrated in the thighs, whose muscle-tendon system is responsible for extending the hip and knee during explosive take-off movements. To jump, a large amount of power needs to be generated at the proximal-most joints to accelerate and propel the body forward (van Ingen Schenau, 1989). During the impulsion phase of jumping, the extensors of the hip and, to a lesser extent, the extensors of the knee are the main contributors to the initial acceleration of the body mass center of a lizard, because the joints of the hindlimb are extended in a proximo-distal sequence through successive extensions of the hip, knee, ankle, and metatarso-tarsal joints, regardless of the performance achieved by a lizard (Legreneur et al., 2012). At last, the tail represented 58% of the headtrunk-tail in both species. This important has a crucial function as a pendulum during jumping by facilitating the substantial rotation of the body through the transverse axis, which often accompanies the airborne phase (Higham et al., 2001; Gillis et al., 2009; Libby et al., 2012). Our results suggest that the morphology of the two ecomorphs differ mainly by a linear scaling factor of the hindlimbs, forelimbs, and head-trunk-tail system. We defined the scaling factors, such as ratios between the absolute values of the and mass of the segments of A. sagrei compared to those of A. carolinensis. The mass scaling factors are 1.18, 0.91, and 0.75 for the hindlimb, forelimb, and headtrunk-tail, respectively. Length scaling factors are 1.02, 1.00, and 0.99 for the hindlimb, forelimb, and head-trunk-tail, respectively. The existence of these correlations is relevant in terms of the possible mechanism of morphological evolution in adaptation to a selective regime created by the interactions between individuals and their environment. Our comparison of the closely related, but differently adapted A. carolinensis and A. sagrei suggests that one of the mechanisms of adaptive evolution may involve relatively slight changes in the proportions of the segments of the body and limb. In conclusion, our comparative analysis of morphometric characteristics of two ecomorphs of anoles creates the basis for a biomechanical model with which to assess the dynamics of locomotion in Anolis and other arboreal squamates. Our analysis also shows that the limbs of the two species are scaled to each other via linear factors. Hence, the observed differences in limb morphometrics in anole species in general may be comparable to the observed differences in beak morphometrics in birds that have undergone adaptive radiation (Wu et al., 2004; Abzhanov et al., 2004; Abzhanov et al., 2006). Given the universality of the growth factors and signaling molecules active during development and growth, it would not be surprising, if early stages of adaptive radiation were driven by similar developmental mechanisms in different organisms. Our results can provide a new methodological approach for comparing jumping behavior and performance in squamates as part of their adaptive radiation. This compartive study focused on adult male individuals of two species of anoles. It follows that the relationship between phenotypic traits and behaviors that we have inferred in this study need to be studied more widely in the clade of anoles to be confirmed (Garland and Adolph, 1994). ACKNOWLEDGMENTS This project was conducted under the ATM program of the Museum National d Histoire Naturelle

(Paris, France) entitled Formes possibles, formes réalisées (2009 2012). LITERATURE CITED MORPHOMETRIC CHARACTERIZATION OF ANOLES 773 Abzhanov A, Kuo WP, Hartmann C, Grant BR, Grant PR, Tabin CJ. 2006. The calmodulin pathway and evolution of elongated beak morphology in Darwin s finches. Nature 442:563 567. Abzhanov A, Protas M, Grant BR, Grant PR, Tabin CJ. 2004. Bmp4 and morphological variation of beaks in Darwin s finches. Science 305:1462. Bels VL, Theys JP, Bennett MR, Legrand L. 1992. Biomechanical analysis of jumping in Anolis carolinensis (Reptilia: Iguanidae). Copeia 2:492 504. Biewener AA, Roberts TJ. 2000. Muscle and tendon contributions to force, work, and elastic energy savings: A comparative perspective. Exerc Sport Sci Rev 28:99. Bock WJ. 1994. Concepts and methods in ecomorphology. J Biosci 19:403 413. Butler MA, King AA. 2004. Phylogenetic comparative analysis: A modeling approach for adaptive evolution. Am Nat 683 695. Cox PG. 1969a. Some aspects of tail regeneration in the lizard, Anolis carolinensis. I. A description based on histology and autoradiography. J Exp Zool 171:127 149. Cox PG. 1969b. Some aspects of tail regeneration in the lizard, Anolis carolinensis. II. The role of the peripheral nerves. J Exp Zool 171:151 159. Crompton RH, Sellers WI. 2007. A consideration of leaping locomotion as a means of predator avoidance in prosimian primates. In: Gursky S, Nekaris K, editors. Primate Anti-Predator Strategies. Springer. p 127. Garland T, Adolph SC. 1994. Why not to do two-species comparative studies: Limitations on inferring adaptation. Physiol Zool 67:797 828. Gillis GB, Bonvini LA, Irschick DJ. 2009. Losing stability: Tail loss and jumping in the arboreal lizard Anolis carolinensis. J Exp Biol 212:604. Harris MA, Steudel K. 2002. The relationship between maximum jumping performance and hind limb morphology/physiology in domestic cats (Felis silvestris catus). J Exp Biol 205(Pt 24):3877 3889. Herrel A, Vanhooydonck B, Porck J, Irschick DJ. 2008. Anatomical Basis of Differences in Locomotor Behavior in Anolis Lizards: A Comparison Between Two Ecomorphs. Bull Mus Comp Zool 159:213 238. Higham TE, Davenport MS, Jayne BC. 2001. Maneuvering in an arboreal habitat: The effects of turning angle on the locomotion of three sympatric ecomorphs of Anolis lizards. J Exp Biol 204(Pt 23):4141 4155. Hoffstetter R, Gasc JP. 1969. Vertebrae and ribs of modern reptiles. In: Gans C, Bellairs A, Parsons TS, editors. Biology of the Reptilia. London: London Academic Press. pp 201 310. Hyman LH. 1992. Hyman s comparative vertebrate anatomy. Chicago: University of Chicago Press. Irschick DJ, Carlisle E, Elstrott J, Ramos M, Buckley C, Van- Hooydonck B, Meyers JAY, Herrel A. 2005a. A comparison of habitat use, morphology, clinging performance and escape behaviour among two divergent green anole lizard (Anolis carolinensis) populations. Biol J Linnean Soc 85:223 234. Irschick DJ, Losos JB. 1998. A comparative analysis of the ecological significance of maximal locomotor performance in Caribbean Anolis lizards. Evolution 52:219 226. Irschick DJ, Vanhooydonck B, Herrel A, Meyers J. 2005b. Intraspecific correlations among morphology, performance and habitat use within a green anole lizard (Anolis carolinensis) population. Biol J Linnean Soc 85:211 222. Irschick DJ, Vitt LJ, Zani PA, Losos JB. 1997. A comparison of evolutionary radiations in mainland and Caribbean Anolis lizards. Ecology 78:2191 2203. Jacobs R, van Ingen Schenau GJ. 1992. Control of an external force in leg extensions in humans. J Physiol 457:611 626. James RS, Navas CA, Herrel A. 2007. How important are skeletal muscle mechanics in setting limits on jumping performance? J Exp Biol 210(Pt 6):923 933. Knouft JH, Losos JB, Glor RE, Kolbe JJ. 2006. Phylogenetic analysis of the evolution of the niche in lizards of the Anolis sagrei group. Ecology 87(sp7):29 38. Legnani G, Righettini P, Zappa B, Casolo F. 1996a. A homogeneous matrix approach to 3D kinematics and dynamics. Part 1: Theory. Mech Mach Theor 31:573 587. Legnani G, Righettini P, Zappa B, Casolo F. 1996b. A homogeneous matrix approach to 3D kinematics and dynamics. Part 2: Applications to chains of rigid bodies and serial manipulators. Mech Mach Theor 31:589 605. Legreneur P, Bels V. Forelimb roles in maximal and submaximal high leaping in squamates (Anolis carolinensis, Voigt, 1832); 2010 30th June 3rd July; Prague, Czech Republic. p 141. Legreneur P, Laurin M, Monteil K, Bels V. 2012. Convergent exaptation of leap up for escape in distantly related arboreal amniotes. Adapt Behav 20:69 79. Libby T, Moore TY, Chang-Siu E, Li D, Cohen DJ, Jusufi A. 2012. Tail-assisted pitch control in lizards, robots and dinosaurs. Nature 481:181 184. Losos JB. 1990a. Ecomorphology, performance capacity, and scaling in West India Anolis lizards: An evolutionary analysis. Ecol Monogr 60:369 388. Losos JB. 1990b. The evolution of form and function: Morphology and locomotor performance in West Indian Anolis lizards. Evolution 44:1189 1203. Losos JB. 1994. Historical contingency and lizard community ecology. In: Vitt LJ, Pianka ER, editors. Lizard Ecology: Historical and Experimental Perspectives. Princeton, NJ: Princeton University Press Princeton. pp 319 333. Losos JB. 2009. Ecology and adaptive radiation of anoles: Lizards in an evolutionary tree. Berkeley and Los Angeles, California: University of California Press.507 p. Losos JB, Creer D, Glossip R, Goellner A, Hampton G, Roberts N, Haskell P, Taylor P, Etling J. 2000. Evolutionary implications of phenotypic plasticity in the hindlimb of the lizard Anolis sagrei. Evolution 54:301 305. Mattingly WB, Jayne BC. 2005. The choice of arboreal escape paths and its consequences for the locomotor behaviour of four species of Anolis lizards. Anim Behav 70:1239 1250. Moermond TC. 1979. Habitat constraints on the behavior, morphology, and community structure of Anolis lizards. Ecology 60:152 164. Montuelle SJ, Daghfous G, Bels VL. 2008. Effect of locomotor approach on feeding kinematics in the green anole (Anolis carolinensis). J Exp Zool Part A Ecol Genet Physiol 309:563 567. Poe S. 2004. Phylogeny of anoles. Herpetological Monographs 18:37 89. Pounds JA. 1988. Ecomorphology, locomotion, and microhabitat structure: Patterns in a tropical mainland Anolis community. Ecol Monogr 58:299 320. Scholz MN, Bobbert MF, Knoek van Soest AJ. 2006. Scaling and jumping: Gravity loses grip on small jumpers. J Theor Biol 240:554 561. Spezzano LC, Jr., Jayne BC. 2004. The effects of surface diameter and incline on the hindlimb kinematics of an arboreal lizard (Anolis sagrei). J Exp Biol 207(Pt 12):2115 2131. Toro E, Herrel A, Irschick D. 2004. The evolution of jumping performance in Caribbean Anolis lizards: Solutions to biomechanical trade-offs. Am Nat 163:844 856. Toro E, Herrel A, Irschick DJ. 2006. Movement control strategies during jumping in a lizard (Anolis valencienni). J Biomech 39:2014 2019. Toro E, Herrel A, Vanhooydonck B, Irschick DJ. 2003. A biomechanical analysis of intra- and interspecific scaling of jumping and morphology in Caribbean Anolis lizards. J Exp Biol 206(Pt 15):2641 2652. van Ingen Schenau GJ. 1989. From rotation to translation: Constraints on multi-joint movements and the unique action of bi-articular muscles. Hum Mov Sci 8:301 337.

774 P. LEGRENEUR ET AL. Vanhooydonck B, Aerts P, Irschick DJ, Herrel A. 2006a. Power generation during locomotion in Anolis Lizards: An ecomorphological approach. In: Herrel A, editor. Ecology and Biomechanics: A Mechanical Approach to the Ecology of Animals and Plants. Boca Raton, Florida, USA: CRC Press. pp 253 269. Vanhooydonck B, Herrel A, Van Damme R, Irschick DJ. 2006b. The quick and the fast: The evolution of acceleration capacity in Anolis lizards. Evolution 60:2137 2147. Vanhooydonck B, Herrel A, Van Damme R, Meyers JJ, Irschick D. 2005. The relationship between dewlap size and performance changes with age and sex in a Green Anole (Anolis carolinensis) lizard population. Behav Ecol Sociobiol 59:157 165. Wells JP, DeMenthon DF. 1987. Measurement of body segment mass, center of gravity, and determination of moments of inertia by double pendulum in Lemur fulvus. Am J Primatol 12:299 308. Williams EE. 1972. The origin of faunas. Evolution of lizard congeners in a complex island fauna: A trial analysis. Evol Biol 6:47 89. Winter DA. 2009. Biomechanics and motor control of human movement, 4th ed. Hoboken, New Jersey: Wiley.370 p. Wu P, Jiang TX, Suksaweang S, Widelitz RB, Chuong CM. 2004. Molecular shaping of the beak. Science 305:1465. Yang SY, Soulé M, Gorman GC. 1974. Anolis lizards of the eastern Caribbean: A case study in evolution. I. Genetic relationships, phylogeny, and colonization sequence of the roquet group. Syst Zool 23:387 399. APPENDIX A: MORPHOMETRIC DATASETS TABLE A1. Morphometric data of Anolis sagrei (N 5 5; mass 5 3.5481 6 0.6170 g; snout-vent 5 51.5 6 3.27 mm) mass/total body mass /Snout-vent Center of mass/ Radius of gyration/ Proximal Distal CoM Proximal Distal Hindlimb 0.062 6 0.002 0.763 6 0.027 0.295 6 0.025 0.705 0.188 6 0.013 0.350 0.730 Thigh 0.036 6 0.002 0.221 6 0.013 0.456 6 0.075 0.544 0.235 6 0.058 0.513 0.592 Leg, tarsus and metatarsus 0.026 6 0.002 0.542 6 0.022 0.320 6 0.031 0.680 0.210 6 0.039 0.383 0.712 Leg 0.018 6 0.001 0.220 6 0.017 0.485 6 0.061 0.515 0.288 6 0.066 0.564 0.591 Tarsus and metatarsus 0.008 6 0.001 0.322 6 0.024 0.328 6 0.022 0.672 0.340 6 0.034 0.472 0.753 Tarsus 0.006 6 0.001 0.123 6 0.003 0.500 6 0.000 0.500 0.289 6 0.000 0.577 0.577 Metatarsus 0.002 6 0.000 0.199 6 0.021 0.500 6 0.000 0.500 0.289 6 0.000 0.577 0.577 Forelimb 0.020 6 0.003 0.466 6 0.038 0.386 6 0.026 0.614 0.177 6 0.048 0.425 0.639 Arm 0.011 6 0.002 0.167 6 0.016 0.477 6 0.077 0.523 0.360 6 0.116 0.598 0.635 Forearm and hand 0.010 6 0.002 0.298 6 0.034 0.410 6 0.013 0.590 0.239 6 0.057 0.475 0.637 Forearm 0.006 6 0.001 0.142 6 0.009 0.500 6 0.000 0.500 0.289 6 0.000 0.577 0.577 Hand 0.003 6 0.001 0.157 6 0.026 0.500 6 0.000 0.500 0.289 6 0.000 0.577 0.577 Head 0.157 6 0.031 0.310 6 0.017 0.615 6 0.036 0.385 0.306 6 0.020 0.687 0.492 Trunk 0.580 6 0.055 0.796 6 0.062 0.489 6 0.040 0.511 0.471 6 0.038 0.679 0.694 Tail 0.098 6 0.025 1.651 6 0.280 0.239 6 0.032 0.761 0.229 6 0.004 0.331 0.795 Head and trunk 0.737 6 0.032 1.106 6 0.056 0.545 6 0.114 0.455 0.589 6 0.021 0.802 0.744 Head, trunk and tail 0.835 6 0.010 2.757 6 0.301 0.586 6 0.103 0.414 0.869 6 0.133 1.048 0.962 Head, forelimbs, trunk and tail 0.876 6 0.004 2.757 6 0.301 0.559 6 0.097 0.441 0.846 6 0.127 1.014 0.954 Abbreviation: CoM 5 Center of mass. The measurements are presented as means 6 standard error. TABLE A2. Morphometric data of Anolis carolinensis (N 5 5; mass 5 4.3016 6 1.1143 g; snout-vent 5 57.4 6 3.89 mm) mass/total body mass /Snout-vent Center of mass/ Radius of gyration/ Proximal Distal CoM Proximal Distal Hindlimb 0.043 6 0.004 0.668 6 0.027 0.290 6 0.021 0.710 0.181 6 0.009 0.346 0.728 Thigh 0.024 6 0.002 0.185 6 0.006 0.414 6 0.084 0.586 0.240 6 0.081 0.516 0.595 Leg, tarsus and metatarsus 0.018 6 0.002 0.482 6 0.022 0.336 6 0.019 0.664 0.224 6 0.045 0.391 0.716 Leg 0.012 6 0.001 0.199 6 0.004 0.488 6 0.047 0.512 0.287 6 0.177 0.563 0.590 Tarsus and metatarsus 0.006 6 0.001 0.283 6 0.020 0.327 6 0.016 0.673 0.261 6 0.154 0.419 0.721 Tarsus 0.004 6 0.001 0.086 6 0.019 0.500 6 0.000 0.500 0.289 6 0.000 0.577 0.577 Metatarsus 0.002 6 0.000 0.197 6 0.004 0.500 6 0.000 0.500 0.289 6 0.000 0.577 0.577 Forelimb 0.018 6 0.001 0.416 6 0.021 0.371 6 0.030 0.629 0.154 6 0.056 0.416 0.633 Arm 0.010 6 0.001 0.151 6 0.010 0.460 6 0.074 0.540 0.398 6 0.195 0.622 0.657 Forearm and hand 0.008 6 0.001 0.266 6 0.019 0.392 6 0.034 0.608 0.301 6 0.119 0.509 0.662 Forearm 0.005 6 0.001 0.117 6 0.025 0.500 6 0.000 0.500 0.289 6 0.000 0.577 0.577 Hand 0.003 6 0.000 0.149 6 0.014 0.500 6 0.000 0.500 0.289 6 0.000 0.577 0.577 Head 0.186 6 0.035 0.339 6 0.021 0.623 6 0.021 0.377 0.866 6 0.085 1.030 0.972 Trunk 0.582 6 0.025 0.739 6 0.028 0.441 6 0.021 0.559 0.889 6 0.095 1.065 0.981 Tail 0.111 6 0.023 1.431 6 0.428 0.248 6 0.028 0.752 0.610 6 0.052 0.818 0.761 Head and trunk 0.768 6 0.027 1.078 6 0.015 0.406 6 0.064 0.594 0.276 6 0.014 0.674 0.474 Head, trunk, and tail 0.879 6 0.009 2.509 6 0.432 0.454 6 0.080 0.546 0.456 6 0.034 0.669 0.685 Head, forelimbs, trunk and tail 0.915 6 0.008 2.509 6 0.432 0.436 6 0.078 0.564 0.230 6 0.019 0.332 0.795 Abbreviation: CoM 5 Center of mass. The measurements are presented as means 6 standard error.

TABLE A3. The mass and of the individual segments are expressed relative to their corresponding body parts, such as the thigh, leg, tarsus, and metatarsus relative to the hindlimb; the arm, forearm, and hand relative to the forelimb; and the head, trunk, and tail relative to the head-trunk-tail system Anolis Thigh Leg Tarsus Metatarsus Arm Forearm Hand Head Trunk Tail Mass/Hindlimb mass Mass/Forelimb mass Mass/Head- Trunk-Tail mass Length/ Hindlimb Length/ Forelimb Length/Head- Trunk-Tail sag. 0.577 6 0.019 car. 0.567 6 0.013 NS 0.294 6 0.024 0.288 6 0.018 NS 0.094 6 0.014 0.094 6 0.013 NS 0.035 6 0.004 0.051 6 0.010 sag. 0.522 6 0.033 car. 0.538 6 0.050 * NS 0.314 6 0.021 0.302 6 0.029 NS 0.164 6 0.019 0.160 6 0.024 sag. 0.188 6 0.038 car. 0.211 6 0.040 sag. 0.289 6 0.014 car. 0.278 6 0.004 NS 0.289 6 0.022 0.299 6 0.010 NS 0.161 6 0.004 0.128 6 0.022 * 0.260 6 0.024 0.295 6 0.011 sag. 0.360 6 0.038 car. 0.362 6 0.024 * NS 0.304 6 0.009 0.280 6 0.055 NS 0.335 6 0.037 0.358 6 0.039 sag. 0.113 6 0.010 car. 0.139 6 0.031 NS NS NS 0.694 6 0.060 0.663 6 0.027 NS 0.291 6 0.037 0.301 6 0.052 NS 0.118 6 0.031 0.126 6 0.027 NS 0.595 6 0.043 0.559 6 0.081 Abbreviations: NS 5 Non Significant; *5 P < 0.05; sag. 5 Anolis sagrei; car. 5 Anolis carolinensis. The measurements are presented as means 6 standard error. NS NS