Is Habitat Preference Associated with Locomotor Performance in Multiocellated Racerunners （Eremias multiocellata） from a Desert Steppe？

Junhuai BI， Yang WANG， Shuran LIand Zhigao ZENG*

1College of Life Sciences， Inner Mongolia Normal University， Hohhot 010022， Inner Mongolia， China

2Key Laboratory of Animal Ecology and Conservation Biology， Institute of Zoology， Chinese Academy of Sciences，Beijing 100101， China

Is Habitat Preference Associated with Locomotor Performance in Multiocellated Racerunners （Eremias multiocellata） from a Desert Steppe？

Junhuai BI1， Yang WANG2， Shuran LI2and Zhigao ZENG2*

1College of Life Sciences， Inner Mongolia Normal University， Hohhot 010022， Inner Mongolia， China

2Key Laboratory of Animal Ecology and Conservation Biology， Institute of Zoology， Chinese Academy of Sciences，Beijing 100101， China

Locomotor performance in lizards is strongly affected by structural habitat. Understanding this relationship allows us to predict species distributions across habitat types. However， little information is available about the ecological role of the locomotion of multiocellated racerunner （Eremias multiocellata） in the desert steppe ecosystem of Inner Mongolia， China. Herein， we studied the effects of habitat structure on the locomotor performance of this lizard species in the fi eld. We found that the sprint speed of this lizard declined signifi cantly with increasing vegetation coverage. Manipulative experiments were further conducted to examine the effects of branch barriers and surface substrates on the sprint speed of the lizard. We found that the sprint speed was signifi cantly infl uenced by the surface substrates and branch barriers， and there were no interactions between them. Branch barriers impeded sprint speed，and E. multiocellata showed better locomotor performance on sandy rather than loamy substrates. Our results indicate that E. multiocellata tends to occupy open areas with sandy substrates， but its locomotor performance is not closely associated with habitat preference.

Habitat， lizard， locomotion， performance， substrate， barrier

1. Introduction

Locomotory abilities of an animal are associated with a variety of ecologically important activities， such as dispersal， mate finding， territory defense， foraging， and escape from predators （Aerts et al.， 2000； Husak et al.，2006； Perry et al.， 2004； Turchin， 1998）. Thus， locomotor performance directly affects the animal's fitness， and is often regarded as a relevant ecological index in fitness measuring （Higham， 2007）. Furthermore， the relationship between habitat type and locomotor performance is vital for understanding fitness and the natural distribution of lizards， because the locomotor performance of lizards is strongly affected by their structural habitat （Irschick and Losos， 1999； Losos and Irschick， 1996）. Therefore，understanding the locomotor performance of lizard species is helpful in predicting their distribution across different habitat types.

The locomotor ability of lizards is correlated to morphological factors， such as body mass， tail， hindlimb，and toes （Borges-Landáez and Shine， 2003； Downes and Shine， 2001； Higham and Russell， 2010； Li C. et al.，2011）， and environmental factors， such as temperature，substrate type， size and incline， and surface diameter（Du et al.， 2005； Gilman and Irschick， 2013； Huang and Tu， 2009； Irschick and Losos， 1999； Li C. et al.，2011）. In addition to morphological factors， structural characteristics of the habitats （e.g. surface substrate diameter and sand grain size） are very important on the locomotor performance （Irschick and Losos， 1999； Li C. et al.， 2011）. As a result， many lizards prefer habitats where they can maximize their locomotor performance. Previous studies have shown that habitat preference is associated with locomotor performance （Higham and Russell， 2010； Irschick and Losos， 1999）. For example，lizards avoid habitats in which sprinting performance is submaximal （Irschick and Losos， 1999）. Microhabitat use in nature may have a profound effect on the locomotor performance of lizards （Vanhooydonck et al.，2005）. However， besides locomotor performance， other ecological factors， including food availability， thermal environment， predation risk， and competition may also affect the habitat preference of lizards （Daly et al.， 2008；Du et al.， 2006； Lisi?i? et al.， 2012； McLaughlin and Roughgarden， 1989； Roches et al.， 2011）. This increases the complexity of the association between habitat preference and locomotor performance.

Lizards are abundant in desert ecosystems worldwide（Pianka， 1986）. However， their distributions are heterogeneous in natural habitats. The multiocellated racerunner， Eremias multiocellata， is a viviparous lacertid lizard， and it generally occupies arid or semi-arid regions covered by sparse vegetation （Zhao et al.， 1999）. It is also a common lizard species in the desert steppe ecosystem of Inner Mongolia， China， and its abundance was strongly affected by the habitat structure （vegetation height， basal cover and bare ground） （Zeng et al.， 2014）. Previous research on this species were primarily focused on variation in life history and the phenotype under different environmental conditions （Li H. et al.， 2011； Tang et al.，2012； Yan et al.， 2011； Zhang et al.， 2010）. However， the effects of habitat structure on locomotor performance and the relationship between locomotor performance and habitat preference have not yet been studied for this species.

The goals of our study were： 1） to evaluate whether lizard movements changed with structural habitats in the field， 2） to determine effects of habitat structure on sprint speed of lizards by manipulating branch barriers and surface substrates in the laboratory， and 3） to test the hypothesis that habitat preference is associated with the locomotor performance of lizards.

2. Materials and Methods

2.1 Field data Our fi eldwork was conducted at the Ordos Key Research Station for Field Observation of Ecological Environments by Sandy Grasslands， China. The station is located in Jungar Banner， a cold semi-arid region of Inner Mongolia （40°12′17′ N， 111°07′43′ E； elevation 1036 m）. The mean annual temperature is 6-7°C， with an average maximum temperature of 39.1°C in July and an average minimum temperature of -32.8°C in January. Annual precipitation ranges from 300 to 380 mm， with most of the rain falling between July and September.

The study area is predominantly sandy grassland with low to moderate levels of sparse vegetation dominated by Artemisia ordosica. Much of the natural habitat in this region have been altered for agricultural purposes and combat desertifi cation. These habitat alterations have increased heterogeneity since the early 2000s. Compared to the natural habitat， the altered habitat is generally either vegetatively： 1） sparser and dominated by Ephedra sinica，or 2） more dense and dominated by Hedysarum leave. In total， three distinct habitat types with different vegetation coverage （sparse， natural， and dense） were selected in our study （Zeng et al.， 2014）.

A raceway （L × W × H： 300 × 30 × 30 cm3） through the vegetation of the three habitats was set up in the fi eld to evaluate the effect of habitat structure （mainly including surface substrates and branch barriers） on the locomotor performance of field-caught adult lizards. A total of 20 adults E. multiocellata （12 males and 8 females） were used in the locomotor performance test. We measured snout-vent length and tail length with a ruler to 1 mm，brachium length， antebrachium length， metacarpus and fore-foot length， thigh length， crus length， and metatarsus and hind-toe length with a caliper to 0.1 mm. We measured body mass with an electronic scale （KL-928；Dongguan Sincerity and Success Electronic Manufactory，Guangdong， China） to 0.01 g at the beginning of the test， and then placed the lizards in individual wooden containers （L × W × H： 55 × 32 × 38 cm3） and left them undisturbed for at least 30 min to acclimate to the ambient temperature prior to each trial. The locomotor performance of lizards is dependent on body temperature，but it is difficult， if not impossible， to precisely control the body temperature of lizards in a fi eld experiment. To minimize temperature effects， we initialized all trials at 1400 and finished before 1600 on sunny days from 20-28 July 2012， during which the fluctuations in ambient temperatures were relatively small. In addition， we measured the body temperature of each lizard just before the locomotor performance test so that we could account for any influence in our statistical models. We assessed locomotor performance by chasing the lizards along the raceway with a paintbrush. Each lizard was tested twice with a rest period of at least 30 min between two trials. The running speed was calculated by dividing the length of racetrack （3 m） with the cumulative time taken by lizards to cross the end of the raceway. The fastest speed for each individual was then used to evaluate the locomotor capability of each lizard in each habitat. We tested for the effects of body temperature on sprint speed using Pearson's correlation test. Correlationanalysis indicated that the sprint speed of lizards was not signifi cantly related to their body temperature （r = -0.031，P = 0.816， n = 60）. Accordingly， body temperature was not included in our statistical model to test for the amonghabitat difference in the sprint speed of lizards.

2.2 Controlled experiments In September 2014， we collected another 10 adult E. multiocellata （4 males and 6 females） from the fi eld and transferred them back to the Institute of Zoology， Chinese Academy of Sciences， for experiments. The lizards were housed in plastic terraria（L × W × D： 600 × 300 × 400 mm3） filled with 50 mm of moist sand. These terraria were kept in a room with a temperature of 24 ± 1°C and a 12 h：12 h light：dark cycle. Food （mealworm， Tenebrio molitor， and crickets， Acheta domestica， dusted with mixed vitamins and minerals）and water were provided ad libitum for the adult lizards. The experiment was performed within three weeks after capturing. Body variables of these lizards were measured at the beginning of the experiment by using the aforementioned methods.

Experimental trials were conducted using a wood runway （L ×W × H： 150 × 20 × 20 cm3）， with a transparent glass panel for lateral filming. The bottom of the runway was covered by 5 cm soil substrate with two treatments （sandy substrate and loamy substrate）. The sandy substrate was derived from sandy soil on the surface of the sparse habitat， and the loamy substrate was from sandy loam on the surface of the dense habitat. The surface soil from the natural habitat was not considered because it was similar to that from the sparse habitat. In addition， a barrier gradient， ranging from no， low， to high branch barriers， was designed by placing some chopsticks on the substrate. The chopsticks were distributed in diamond shapes with 30 cm long diameter and 4 cm or 8 cm short diameter. The former with 4 cm short diameter and the latter with 8 cm short diameter were regarded as treatments of high and low branch barriers，respectively. Therefore， we used a total of six treatments（2 surface substrates × 3 branch barriers） in our controlled experiments.

We quantified the locomotor performance （sprint speed） of the lizards by racing each individual down the racetrack （marked at 20 cm intervals） and encouraged with a soft paintbrush. Individuals were raced 3 times for each treatment at 30 ± 1°C， with 1 h of rest between each consecutive race. We recorded the running with a video camera from a lateral perspective. For each individual in each treatment， we used the fastest 20 cm from each of the three races to calculate sprint speed and quantify locomotor performance.

2.3 Data analysis One-way repeated-measures analysis of variance （ANOVA） was used to determine the amonghabitat difference in the sprint speed of E. multiocellata with sex as the between-subjects factor and sprint speed in each habitat as the repeated measure. Two-way repeatedmeasures ANOVA was used to test the effects of surface substrates and branch barriers on the sprint speed. Partial correlation analyses was used to test for the effects of body variables on the sprint speed of lizards with habitat type in the field data and surface substrates and branch barriers in the experiment data as control variables. To remove the effects of some body variables on locomotor performance， univariate analysis of covariance（ANCOVA） was used to determine differences among habitats （or among surface substrates and branch barriers）in the sprint speed with body variables as the covariates. All data were analyzed by the software package of SPSS 19.0.

3. Results

The sprint speed of the lizards differed among habitats（F2，36= 29.58， P ＜ 0.0001）， and a signifi cant interaction was not found between habitat and sex （habitat * sex：F2，36= 0.03， P = 0.967）， indicating that E. multiocellata performed differently in each habitat. Partial correlation analyses indicated that the sprint speed of the lizards in the field was significantly related to body mass （r = 0.499， P ＜ 0.0001， n = 57）， snout-vent length （r = 0.437，P = 0.001， n = 57）， brachium length （r = 0.504， P ＜0.0001， n = 57）， and thigh length （r = 0.474， P ＜ 0.0001，n = 57）. After the effects of these four body variables on locomotor performance had been statistically removed，the sprint speed significantly declined with increasing vegetation coverage in the habitats （F2，53= 24.67， P ＜0.0001） （Figure 1）.

Surface substrates and branch barriers had apparent effects on the locomotor performance of E. multiocellata in the controlled experiments （Figure 2）. The sprint speed of the lizards decreased by branch barriers （F2，18= 12.81， P ＜ 0.001）， and the lizards had faster sprint speed on the sandy than loamy substrates （F1，9= 16.46， P = 0.003）. The interaction between the surface substrates and branch barriers had no signifi cant effect on the sprint speed （surface substrates * branch barriers： F2，18= 1.20，P = 0. 324）. Partial correlation analyses indicated that the sprint speed of the lizards in the controlled conditions was signifi cantly related to snout-vent length （r = 0.315，P = 0.016， n = 56） and crus length （r = -0.296， P = 0.024，n = 56）. After the effects of these two body variables onlocomotor performance had been statistically removed，the sprint speed significantly decreased with increasing branch barriers （F2，52= 8.49， P = 0.001）， and the sprint speed was still faster on the sandy than loamy substrates（F1，52= 20.36， P ＜ 0.0001）. E. multiocellata showed better locomotor performance on sandy rather than loamy substrates （Figure 2）.

4. Discussion

In this study， we measured the ecological performance of multiocellated racerunners moving through their natural environment and highlighted the differences in their locomotor performance in different structural habitats. We found that the locomotor performance of E. multiocellata declined markedly with increasing vegetation coverage in the structural habitats. Our experiment results indicated that surface substrate and branch barrier are two important components of habitat structure influencing the lizard's locomotor performance. Previous studies have shown that natural habitats had higher abundance of E. multiocellata than structurally sparse and dense vegetation （Zeng et al.，2014）. Obviously， our studies both in the field and laboratory supported that locomotor performance of E. multiocellata is not closely associated with habitat preference， despite the significant effects of habitat structure on locomotion. Similar results were found in two sand lizard species （Uma scoparia and Callisaurus draconoides） in the Mojave Desert， their differences in habitat distribution are not closely related to locomotor performance （Korff and McHenry， 2011）. Furthermore，our experiments show that E. multiocellata performs better on sandy rather than on loamy substrates. Thus， this lizard species is expected to occupy primarily open areas with sandy substrates.

The locomotor performance measured under our controlled conditions was very close to that measured under natural conditions for E. multiocellata in our studies. Field speeds of E. multiocellata in the sparse，natural， and dense habitats were about 92%， 95% and 108% of controlled trackway speeds that measured under corresponding controlled conditions， respectively（Figures 1， 2）. In contrast， larger discrepancies in locomotor performance usually occur between fi eld- and laboratory-based measures in other lizards. For example，some lizards have been recorded in the field to run at about 78%-90% of the maximum speed measured under controlled experiments （Higham and Russell， 2010；Irschick and Losos， 1998）， whereas others run at higher（about 118%） velocity in the fi eld than in the laboratory（Higham and Russell， 2010； Jayne and Ellis， 1998）. Therefore， we concluded that maximum performance was accurately measured under our controlled conditions for E. multiocellata. More importantly， the pattern of habitat effects on locomotor performance was the same in the fi eld and laboratory tests， suggesting that the relationship between locomotor performance and habitat structure is relative constant with little effects from other ecologicalfactors. The consistency of our results from the field and laboratory tests provides solid evidence to test for the hypothesis that habitat preference is associated with locomotor performance in lizards.

Previous studies suggest that habitat structures infl uence the locomotor performance of lizards （Du et al.，2005； Irschick and Losos， 1999）. Some attributes of habitat structure may impede locomotion when lizards move through their natural environment. Structurally，habitats become more complex with increasing vegetation coverage， so lizards may frequently encounter more branch barriers located along their traveling routes. Consequently， lizard movements change with structural habitats in the field， and E. multiocellata obviously showed an impeded locomotor performance in a denser habitat. This may be an important factor that influences its distribution in a dense habitat where E. multiocellata is rarely found （Zeng et al.， 2014）， because some lizards avoid structural habitats in which their maximal sprinting abilities are impaired （Irschick and Losos， 1999）. However， a great mismatch between the locomotor performance and population density of the lizards occurred in the field study. E. multiocellata is more abundant in natural rather than dense habitats （Zeng et al.， 2014）， but the locomotor performance was similar between these two habitats. E. multiocellata is more abundant in natural rather than sparse habitats （Zeng et al.， 2014）， but the locomotor performance was better in the sparse than natural habitats. These results suggest that locomotor performance is not closely associated with habitat preference in this lizard species. Other than the effects of locomotor performance， ecological factors such as soil， temperature， predators， competition， and prey may all interact to affect the distribution of lizards（Daly et al.， 2008； Du et al.， 2006； Lisi?i? et al.， 2012；McLaughlin and Roughgarden， 1989； Sun et al.， 2014；Zaady and Bouskila， 2002）. The habitat preference of E. multiocellata may be associated with some other ecological factors to a greater extent. For example，we found that E. multiocellata had a higher predation risk from birds in sparse habitats than in dense habitats（unpublished data）. Habitat structure affects predation risk （López and Martín， 2013）. This lizard species is especially vulnerable to predators while moving across relatively open habitats. Therefore， sparse habitats， where E. multiocellata showed the best locomotor performance，are not dominantly occupied by the lizard species.

Locomotor performance depends on a mechanical interaction between the feet of the lizard and the surface substrate. Effects of this interaction may be substrate-specific and species-specific because of their different locomotor morphologies. For example，Callisaurus draconoides without toe fringes ran with a faster maximum speed than Uma scorparia with toe fringes on dune sand， but not on wash sand （Korff and McHenry， 2011）. This effect was also substrate-specific for E. multiocellata in our experiments. We found that the locomotor performance of this lizard species was significantly different among surface substrates from different structural habitats， and the surface substrates had no signifi cantly interactive effects with the branch barriers. Substrate type affects the maximum instantaneous speed of animals： the gecko Hemidactylus garnoti attains higher maximum instantaneous speeds on wood surfaces than on metal or cloth surfaces （Vanhooydonck et al.， 2005）. For the sympatric toad-headed lizard Phrynocephalus frontalis in the desert steppe， substrate （i.e.， sand grain size） also has signifi cant effects on its running speed， and it is a factor that infl uences its habitat selection （Li C. et al.， 2011）. Obviously， locomotor ability is important for a lizard species， especially when escaping from predators or capturing prey （Irschick and Losos， 1999； Pounds， 1988）. When natural selection acts on locomotor performance，substrate choice of the lizard species could infl uence its fitness. Consequently， lizards may dominantly use and inhabit a preferred sandy substrate with a locomotory advantage. Therefore， only a few E. multiocellata are distributed in dense habitats with loamy substrates （Zeng et al.， 2014）.

Effects of habitat structure， especially habitat substrates and physical structure， on locomotor performance have been extensively studied and well documented in lizards （Irschick and Losos， 1998； Mattingly and Jayne，2004； Mcelroy et al.， 2007； Vanhooydonck et al.， 2005；Vanhooydonck and Van Damme， 2003）. However， the ability of these animals to move through branch barriers has received much less attention. Animals often encounter different barriers along their traveling routes in their natural environment. The ability of reptiles to negotiate different barriers is likely to be under strong selection because habitats with dense vegetation impose spatial challenges for ecologically relevant activities， such as foraging and escape from predators （Irschick and Losos，1999； Pounds， 1988）. Therefore， understanding the effects of habitat barriers on the locomotion of lizards is very important.

Working on the relationship between maximal locomotor performance of lizards and the obstacles of habitat structures such as perch height and incline，some scientists found that locomotor speed signifi cantlydecreases with increasing perch height or when lizards run on steep inclines （Irschick and Jayne， 1998； Jayne and Irschick， 2000； Mcelroy et al.， 2007）. Kohlsdorf and Biewener （2006） also studied the kinematics of lizard locomotion when animals move over obstacles encountered along traveling routes. However， few studies have considered the effects of branches as physical barriers in structural habitats. Our experiments showed that branch barriers signifi cantly impeded the locomotor performance of E. multiocellata. Lower sprint speeds when negotiating branch barriers are likely associated with a movement deceleration to evade the obstacle.

Acknowledgements We sincerely appreciate W. G. DU for his help during the experiment and his suggestions for the manuscript. Our study was performed under the approval from the Animal Ethics Committee at the Institute of Zoology， Chinese Academy of Sciences（IOZ14001）.

Aerts P.， Van Damme R.， Vanhooydonck B.， Zaaf A.， Herrel A. 2000. Lizard locomotion： how morphology meets ecology. Netherlands J Zool， 50： 261-277

Borges-Landáez P. A.， Shine R. 2003. Infl uence of toe-clipping on running speed in Eulamprus quoyii， an Australian scincid lizard. J Herpetol， 37： 592-595

Daly B. G.， Dickman C. R.， Crowther M. S. 2008. Causes of habitat divergence in two species of agamid lizards in arid central Australia. Ecology， 89： 65-76

Downes S.， Shine R. 2001. Why does tail loss increase a lizard's later vulnerability to snake predators？ Ecology， 82： 1293-1303

Du W.， Lin Z.， Shou L.， Ji X. 2005. Morphological correlates of locomotor performance in four species of lizards using different habitats. Zool Res， 26： 41-46 （In Chinese）

Du W. G.， Shou L.， Shen J. Y. 2006. Habitat selection in two sympatric Chinese skinks， Eumeces elegans and Sphenomorphus indicus： do thermal preferences matter？ Can J Zool， 84： 1300-1306

Gilman C. A.， Irschick D. J. 2013. Foils of fl exion： the effects of perch compliance on lizard locomotion and perch choice in the wild. Funct Ecol， 27： 374-381

Higham T. E. 2007. The integration of locomotion and prey capture in vertebrates： morphology， behavior， and performance. Integr Comp Biol， 47： 82-95

Higham T. E.， Russell， A. P. 2010. Divergence in locomotor performance， ecology， and morphology between two sympatric sister species of desert-dwelling gecko. Biol J Linn Soc， 101：860-869

Huang S.-P.， Tu M.-C. 2009. Locomotor and Elevational Distribution of a Mountainous Lizard， Takydromus hsuehshanensis， in Taiwan. Zool Stud， 48： 477-484

Husak J. F.， Fox S. F.， Lovern M. B.， Van Den Bussche R. A. 2006. Faster lizards sire more offspring： sexual selection on whole-animal performance. Evolution， 60： 2122-2130

Irschick D. J.， Losos J. B. 1998. A comparative analysis of the ecological significance of maximal locomotor performance in Caribbean Anolis lizards. Evolution， 52： 219-226

Irschick D. J.， Losos J. B. 1999. Do lizards avoid habitats in which performance is submaximal？ The relationship between sprinting capabilities and structural habitat use in Caribbean anoles. Am Nat， 154： 293-305

Irschick D. J.， Jayne B. C. 1998. Effects of incline on speed，acceleration， body posture and hindlimb kinematics in two species of lizard Callisaurus draconoides and Uma scoparia. J Exp Biol， 201： 273-287

Jayne B. C.， Ellis R. V. 1998. How inclines affect the escape behaviour of a dune-dwelling lizard， Uma scoparia. Anim Behav， 55： 1115-1130

Jayne B. C.， Irschick D. J. 2000. A fi eld study of incline use and preferred speeds for the locomotion of lizards. Ecology， 81：2969-2983

Kohlsdorf T.， Biewener， A. A. 2006. Negotiating obstacles： running kinematics of the lizard Sceloporus malachiticus. J Zool， 270：359-371

Korff W. L.， McHenry M. J. 2011. Environmental differences in substrate mechanics do not affect sprinting performance in sand lizards （Uma scoparia and Callisaurus draconoides）. J Exp Biol，214： 122-130

Li C.， Lian X.， Bi J.， Fang H.， Maul T. L.， Jiang Z. 2011. Effects of sand grain size and morphological traits on running speed of toad-headed lizard Phrynocephalus frontalis. J Arid Environ， 75：1038-1042

Li H.， Qu Y. F.， Ding G. H.， Ji X. 2011. Life-history variation with respect to experienced thermal environments in the lizard，Eremias multiocellata （Lacertidae）. Zool Sci， 28： 332-338

Lisi?i? D.， Drakuli? S.， Herrel A.， ?iki? D.， Benkovi? V.， Tadi?，Z. 2012. Effect of competition on habitat utilization in two temperate climate gecko species. Ecol Res， 27： 551-560

López P.， Martín J. 2013. Effects of microhabitat-dependent predation risk on vigilance during intermittent locomotion in Psammodromus algirus lizards. Ethology， 119： 316-324

Losos J. B.， Irschick D. J. 1996. The effects of perch diameter on the escape behavior of Anolis lizards： laboratory-based predictions and fi eld tests. Anim Behav， 51： 593-602

Mattingly W. B.， Jayne B. C. 2004. Resource use in arboreal habitats： structure affects locomotion of four ecomorphs of Anolis lizards. Ecology， 85： 1111-1124

Mcelroy E. J.， Meyers J. J.， Reilly S. M.， Irschick， D. J. 2007. Dissecting the effects of behaviour and habitat on the locomotion of a lizard （Urosaurus ornatus）. Anim Behav， 73： 359-365

McLaughlin J. F.， Roughgarden J. 1989. Avian predation on Anolis lizards in the Northeastern Caribbean-an inter-island contrast. Ecology， 70： 617-628

Perry G.， LeVering K.， Girard I.， Garland Jr. T. 2004. Locomotor performance and social dominance in male Anolis cristatellus. Anim Behav， 67： 37-47

Pianka E. R. 1986. Ecology and natural history of desert lizards. Princeton： Princeton University Press

Pounds A. J. 1988. Ecomorphology， locomotion， and habitat structure： patterns in a tropical mainland Anolis community. Ecol Monogr， 58： 299-320

Roches S. D.， Robertson J. M.， Harmon L. J.， Rosenblum， E. B. 2011. Ecological release in White Sands lizards. Ecol Evol， 1：571-578

Sun B.， Tang W.， Zeng Z.， Du W. 2014. The seasonal acclimatisation of locomotion in a terrestrial reptile， Plestiodon chinensis （Scincidae）. Asian Herpetol Res， 5： 197-203

Tang X. L.， Yue F.， Yan X. F.， Zhang D. J.， Xin Y.， Wang C.，Chen Q. 2012. Effects of gestation temperature on offspring sex and maternal reproduction in aviviparous lizard （Eremias multiocellata） living at high altitude. J Therm Biol， 37： 438-444

Turchin P. 1998. Quantitative analysis of movement： measuring and modeling population redistribution in animals and plants. Sunderland： Sinauer Associates Inc

Vanhooydonck B.， Andronescu A.， Herrel A.， Irschick， D. J. 2005. Effects of substrate structure on speed and acceleration capacity in climbing geckos. Biol J Linn Soc， 85： 385-393

Vanhooydonck B.， Van Damme， R. 2003. Relationships between locomotor performance， microhabitat use and antipredator behaviour in lacertid lizards. Funct Ecol， 17： 160-169

Yan X. F.， Tang X. L.， Yue F.， Zhang D. J.， Xin Y.， Wang C.，Chen Q. 2011. Influence of ambient temperature on maternal thermoregulation and neonate phenotypes in a viviparous lizard，Eremias multiocellata， during the gestation period. J Therm Biol，36： 187-192

Zaady E.， Bouskila， A. 2002. Lizard burrows association with successional stages of biological soil crusts in an arid sandy region. J Arid Environ， 50： 235-246

Zeng Z. G.， Bi J. H.， Li S. R.， Chen S. Y.， Pike D. A.， Gao Y.， Du W. G. 2014. Effects of habitat alteration on lizard community and food web structure in a desert steppe ecosystem. Biol Conserv， 179： 86-92

Zhang D. J.， Tang X. L.， Yue F.， Chen Z.， Li R. D.， Chen Q. 2010. Effect of gestation temperature on sexual and morphological phenotypes of offspring in a viviparous lizard， Eremias multiocellata. J Therm Biol， 35： 129-133

Zhao E. M.， Zhao K. T.， Zhou K. Y. 1999. Fauna Sinica， Reptilia Vol. 2， Squamata. Beijing： Chinese Science Press

Dr. Zhigao ZENG， from the Institute of Zoology，Chinese Academy of Sciences， Beijing， China， with his research focusing on ecological adaptation and conservation biology of reptiles.

E-mail： zengzhg@ioz.ac.cn

17 December 2014 Accepted： 11 May 2015