Ecomorphological Differences in Toepad Characteristics Between Mainland and Island Anoles
Total Page:16
File Type:pdf, Size:1020Kb
52 M.-O. ROÈ DEL ET AL. MHNG 1034.14±19, Foulassi and surrounding, Cam- 22 November 1990, coll. Largen.; Democratic Republic eroon, 1960±1961, coll. Perret; MNHN 1959.0025± of Congo (Zaire): ZSM 44/1984, Prov. Kivu, Irangi at 0026, Cameroun, coll. Perret; MRAC B.103114±103116, river Luoho, 850 m, Zaire, 4.-7. March 1984, coll. Foulassi, 1960, coll. Perret; Congo: LIV 1991.11.208 Fuchs; MRAC B.112549 Bunyakiri, territory of Kalehe, Goumina, Congo Brazzaville, 27 November 1990; LIV Kivu, Congo Belge, 5.-7. June 1949, coll. Laurent. 1991.11.206±207 (tadpoles), BeÂna, Congo Brazzaville, Acanthixalus sonjae sp. nov.Ðtype series. Journal of Herpetology, Vol. 37, No. 1, pp. 52±58, 2003 Copyright 2003 Society for the Study of Amphibians and Reptiles Ecomorphological Differences in Toepad Characteristics between Mainland and Island Anoles THOMAS E. MACRINI,1,2 DUNCAN J. IRSCHICK,3 AND JONATHAN B. LOSOS1 1Department of Biology, Campus Box 1137, Washington University, Saint Louis, Missouri 63130-4899, USA 3Department of Ecology and Evolutionary Biology, 310 Dinwiddie Hall, Tulane University, New Orleans, Louisiana 70118, USA ABSTRACT.ÐAnoles (Anolis, Chamaeleolis, Chamaelinorops, Phenacosaurus) differ greatly in size and shape of their subdigital toepads. We sampled 79 anole species to address two issues. First, is toepad size signif- icantly related to habitat use? Second, do anoles from mainland Central and South America differ from Caribbean anoles in either the size of their toepads or in the relationship between toepad shape and habitat use? Among all anoles, toepad size increases with both perch height and diameter, although when the effects of body size are removed, these relationships are weakened. Morphological differences exist between toe- pads of anoles from the two regions. Caribbean anoles have larger toepads relative to snout±vent length and wider pads at larger body sizes compared to mainland species. Relationships between toepad mor- phology and habitat use characteristics do not differ signi®cantly between the two groups of anoles. How- ever, with effects of size removed, Caribbean anoles have larger and wider toepads relative to both perch height and diameter. These ®ndings suggest that anoles in island habitats may require larger toepads than mainland anoles to use their arboreal habitats. The potential functional consequences of these differences in toepad morphology provide a promising avenue for future studies. A fundamental issue in evolutionary biology point of comparison for understanding how en- concerns factors that in¯uence shape of adaptive vironmental differences shape the pathways of radiations. One means of addressing this issue adaptive radiations. In both the Caribbean and is to compare morphological and ecological the mainland, anoles have diversi®ed tremen- characteristics of closely related groups of spe- dously (. 140 species in each region) under dif- cies in different environments. An obvious com- ferent selective pressures, producing species parison is between mainland and island faunas, that vary in morphology, habitat use, behavior, which often display a remarkable array of dif- and function (Williams, 1972, 1983; Fitch, 1975, ferences in life history, morphology, and behav- 1976; Andrews, 1979; Pounds, 1988; Losos, ior (e.g., Nussbaum and Wu, 1995; Case and 1995). In a comparison of 16 mainland and 33 Schwaner, 1993; Petren and Case, 1997). Yet, Caribbean anole species, Irschick et al. (1997) with a few exceptions, most of these studies showed that ecomorphological relationships dif- have focused only on one or two species and fered dramatically in the two regions. An exten- have not quantitatively compared groups of sion of that analysis, presented here, focuses on species in mainland and island habitats. differences among regions in a key morpholog- Caribbean and mainland (Central and South ical characteristic, the subdigital toepad. American) anoline lizards provide an excellent Lizard toepads provide an excellent oppor- tunity to study patterns of adaptation and mac- 2 Corresponding Author. Present address: Jackson roevolution (Bauer and Good, 1986; Irschick et School of Geosciences, University of Texas at Austin, al., 1996). Previous work has shown that toepads Austin, Texas 78712-0254, USA; E-mail: tmacrini@ with clinging capabilities evolved independent- mail.utexas.edu ly in three different groups of arboreal lizards: TOEPAD DIFFERENCES IN MAINLAND AND ISLAND ANOLES 53 geckos, anoles, and some skinks (Ruibal and MATERIALS AND METHODS Ernst, 1965; Williams and Peterson, 1982). In Morphological Data.ÐFive hundred ®fty-six each case, the expanded subdigital pad is divid- preserved specimens of 79 species of anoles ed into a number of transverse scales, termed were examined (63 from the Caribbean, 16 from lamellae. Each lamella is covered with tens of Central or South America; list of species and thousands of microscopic, hairlike keratinous means of morphological measurements avail- structures called setae, with each containing able upon request). Included in this sample hundreds of projections that terminate in spat- were species of Chamaeleolis, Chamaelinorops,and ula-shaped structures (Maderson, 1964; Ruibal Phenacosaurus, all of which have arisen from and Ernst, 1965; Bauer and Good, 1986). These within Anolis (Jackman et al., 1999). spatulae play a prominent role in the ability of Five morphological measurements were tak- these lizards to cling to smooth surfaces (Hiller, en, all on adult males: snout±vent length (SVL), 1968; Cartmill, 1985; Irschick et al., 1996; Au- and width and area of the toepad on the fourth tumn et al., 2000). A recent study that measured toe of the fore- and hind foot. The fourth toe the adhesive force of setae of the Tokay gecko was chosen because it is usually the largest (Gekko gecko) supports the hypothesis that Van among anoles. Toepad area was measured by der Waal's forces may be responsible for adhe- digitizing the circumference of the subdigital sion (Autumn et al., 2000). toepad (not including the claw and the rest of Repeated evolution of toepads in arboreal liz- the digit) using a video imaging system (Mor- ards and the obvious functional advantage they phosys; Meacham and Duncan, 1990) connected confer leaves little doubt that these structures to a personal computer. Toepad width was mea- evolved as adaptations for climbing. Some au- sured to 0.1 mm with calipers across the widest thors suggest that toepads are key innovations part of the pad. Each of the measurements was that facilitated occupation of a new adaptive taken from the ventral aspect on the right side zone and subsequent evolutionary diversi®ca- of the lizard. tion (e.g., Russell, 1979; Larson and Losos, Habitat Use Data.ÐTwo measures of habitat 1996). However, surprisingly little research has use shown to be important for segregating anole directly examined the relationship between toe- species (Rand, 1964; Schoener, 1968; Losos, pad characteristics and habitat use, even though 1990b; Losos and Irschick, 1996) were employed such information is necessary to fully test hy- here: mean perch height (PH) and mean perch potheses about the role of toepads in adaptive diameter (PD; data from Losos, 1990b; Irschick diversi®cation. and Losos, 1996; Losos and Irschick, 1996; Losos Anolis lizards differ greatly in the size and and de Queiroz, 1997). Both habitat features structure of their toepads and also in their de- have the potential to in¯uence toepad size and gree of arboreality. Researchers have noted that shape. For instance, lizard species that perch species with more lamellae perch higher off the higher than other species might bene®t from ground than species with fewer lamellae (Col- having relatively larger toepads (thus allowing lette, 1961; Williams, 1983; Glossip and Losos, more sure-footedness) because the cost of fall- 1997). However, other aspects of the toepad ing to the ground is greater. Also, relatively larg- have received less attention. In particular, all er toepads may be useful for grasping narrow else equal, one would expect relatively larger branches by providing greater ¯exibility and toepads to have more setae and thus provide thus permitting the pads to better wrap around greater clinging ability than relatively smaller the surface. Consequently, one might expect a toepads (Peterson, 1983; Bauer and Good, 1986) negative relationship among species between and, indeed, among a sample of anoles, geckos, preferred perch diameter and toepad size. and skinks, a correlation exists between relative Statistical Analyses.ÐMean values for each toepad size and clinging ability (Irschick et al., variable for each species were used in all anal- 1996). As a result, one might expect more ar- yses. These means were natural log-trans- boreal species to have relatively larger toepads formed prior to analysis. Ecological data were than more terrestrial species. not available for all species; therefore, sample Here, we addressed the above issues by fo- sizes for morphology-ecology comparisons were cusing on the following questions: (1) How is smaller than for morphology-morphology com- toepad size related to habitat use? (2) Do main- parisons. land and Caribbean anoles differ in toepad Most quantitative and ecological variables in- shape and in the relationships between toepad crease with body size both within and among size and habitat use? To answer these questions, species of anoles (Losos, 1990b; Macrini and Ir- we measured the toepad characteristics of 79 schick, 1998). Because toepads are important for species of anoles and compared these data to clinging to vertical surfaces, mass is an obvious previously collected data on habitat use. index of body size, but the use of mass is prob- 54 T. E. MACRINI ET AL. TABLE 1. Results from analyses of covariance relating toepad size to body size between mainland and Caribbean anole species. Abbreviations: ff 5 forefoot, hf 5 hind foot, * P , 0.05, ** P , 0.01, *** P , 0.001. Test of heterogeneity Test of heterogeneity Relationship of slopes of intercepts Toepad width (ff) vs. SVL F1,75 5 4.09* Ð Toepad width (hf) vs.