ANIMAL BEHAVIOUR, 1999, 58, 77–83 Article No. anbe.1999.1132, available online at http://www.idealibrary.com on Laboratory endurance capacity predicts variation in field locomotor behaviour among lizard species THEODORE GARLAND, JR Department of Zoology, University of Wisconsin-Madison (Received 23 July 1998; initial acceptance 14 November 1998; final acceptance 22 March 1999; MS. number A8260) I measured locomotor endurance capacities of lizards on a motorized treadmill in the laboratory and compared average values for different species with quantitative measures of their movement in the field (percentage of time moving, N=15 species; moves/min, N=13; daily movement distance, N=11). I hypothesized that endurance would be positively related to all three movement indices. Relationships between log endurance and log movement were computed as conventional Pearson product–moment correlations and as the equivalent with phylogenetically independent contrasts. Endurance was signifi- cantly positively related to both the percentage of time moving and the daily movement distance. This is the first study to demonstrate such relationships with phylogenetically based statistical methods. These results suggest that endurance capacities of lizards are coadapted with their typical locomotor behaviour.  1999 The Association for the Study of Animal Behaviour Cross-species comparisons can provide important infor- mation about evolutionary patterns and processes (Harvey & Pagel 1991) and have a long history in animal behaviour (Hailman 1998). In particular, correlations between phenotypic traits and ecological or environ- mental variables provide evidence of adaptation; that is, that natural selection has played a role in causing the phenotypic differences among species (Garland & Adolph 1994; Doughty 1996). In addition to adaptation of single traits, different aspects of the phenotype are generally expected to show coadaptation. For example, species that show high movement rates should have high stamina, whereas those that sprint frequently should be able to attain high maximal speeds (Huey et al. 1984; Hertz et al. 1988; Irschick & Losos 1990; Losos 1990a, b; Pough & Taigen 1990; Pough et al. 1992; Garland 1993, 1994; Perry 1999). Ecological and evolutionary physiologists and mor- phologists have provided many examples of cross-species correlations between morphology or physiology on the one hand and behaviour or ecology on the other (Feder et al. 1987; Pough et al. 1992; Wainwright & Reilly 1994), thus providing evidence of evolutionary adaptation. Tra- ditionally, most studies have involved morphological or physiological traits at levels below the whole animal, such as limb proportions, bill dimensions, blood charac- teristics or enzyme activities. A weakness of these studies is that they neglect the crucial intermediate phenotype of maximal whole-animal performance abilities (Arnold 1983; Pough 1989). Hence, many recent studies have attempted to include direct measures of organismal performance (e.g. Jayne & Ellis 1998). Theoretically, measures of organismal performance should be more direct targets of natural selection and hence should cor- relate more strongly with behavioural ecology than would lower-level traits (Garland & Carter 1994; Garland & Losos 1994). Here, I tested the hypothesis that a laboratory measure of stamina can predict movement behaviour of lizards in the field. Following several pre- vious studies (e.g. Huey et al. 1984; Pough et al. 1992; Garland 1993, 1994; references therein), I hypoth- esized that measures of field locomotor behaviour would correlate positively with endurance. Lizards are diverse in form, behaviour and ecology (Greer 1989; Vitt & Pianka 1994) and have served as a particularly common model in comparative studies of locomotor performance (Garland & Losos 1994; Gans et al. 1997). Various aspects of locomotor ability have been measured in the laboratory, including clinging and jumping (Losos 1990a, b; Irschick et al. 1996). Sprinting ability has been especially commonly studied, usually by measurement on a photocell-timed racetrack (e.g. Huey et al. 1984; Losos 1990a, b; Miles 1994a; Bauwens et al. 1995; Zani 1996; Bonine & Garland, in press). Measures of locomotor endurance (stamina) have been less commonly compared among species of lizards (Bennett 1980; Garland 1993; see also Miles 1994b on population variation). Cullum (1997) compared sprint speed, maximal exertion (distance run to exhaustion Correspondence: T. Garland, Department of Zoology, 430 Lincoln Drive, University of Wisconsin, Madison, WI 53706-1381, U.S.A. (email: tgarland@macc.wisc.edu). 0003–3472/99/070077+07 $30.00/0  1999 The Association for the Study of Animal Behaviour 77