Teratoscincus roborowskii is endemic to sand dune habitats in the Turpan Basin, Western China. Its home range was investigated using mark-recapture techniques from June 2008 to May 2009 and compared between the reproductive season (RS, June 2008 to August 2008) and the non-reproductive season (NRS, from September 2008 to May 2009 excluding the 5-6 month period of hibernation). In total, 283 individuals were marked by toe-clipping, of which 24 individuals were recaptured more than three times in the RS, and 43 in the NRS (13 were recaptured in both seasons) and were used to quantify home range size. Their minimum convex polygon (MCP) ranges were calculated using MapGis software and analyzed with respect to the effects of factors such as sex, season, and body size. There were significant differences in home range size among male adults, female adults, and juveniles in the NRS, whereas, in the RS, the three groups were not significantly different. There were no significant differences in home range size between the sexes, in either the RS or the NRS. Home range and snout-vent length (SVL) were significantly correlated in the NRS, and home range size and SVL of adults were significantly correlated throughout the whole year (combining RS and NRS data). In the NRS, the home range sizes of adults were significantly larger than that of juveniles. There were significant differences in the home ranges of adults between the RS and the NRS (U=41, P=0.046) but, in juveniles, there was no difference (U=159, P=0.537). There was no difference between sexes in home range size but body size affected the home range. Home range size was correlated with body size, presumably, because of energetic requirements. The absence of a difference between sexes in the size of the home range of T. roborowskii is consistent with the absence of significant sexual dimorphism in body size in this species, and by the need for adult males to establish an optimal habitat. However, chases and intrusions by males were rarely observed and, because the eggs of pregnant females were contained in a relatively small abdominal cavity, the costs of reproduction during the RS were low for both adult males and females. Seasonal shifts in home range size appeared to have complex effects on its temporal-spatial usage (e.g. patterns of daily activity, thermoregulation, and food availability). Since T. roborowskii is a strictly nocturnal reptile, the pattern of daily activity and thermoregulation varied seasonally. Overall, seasonal differences of home range were non-significant but seasonal changes in the reproductive status of many adults led to significant seasonal differences of home range between the RS and the NRS. Juveniles were sexually immature and this was reflected in their non-significant seasonal differences of home range. Food availability exerted significant seasonal effect on adults. Adults occupying preferred areas, which are better for courtship, mating and food requirements, forced juveniles into less favorable areas during the RS. This meant that juveniles foraged over larger areas in the RS than that in the NRS. As the density of the population was increased in the NRS by the birth of new juveniles, this might intensify the trend of widening home range for juveniles. However, there was no significant difference in the home ranges of juveniles at different seasons. In conclusion, the most important factors that determine the home range size of T. roborowskii are seasonal variations in the type and availability of food resources, and reproductive status.