Soil respiration is not only an indicator of belowground metabolic activity of roots and soil microbes, but also a necessary component of carbon cycle. Measurement of the CO2 efflux from soil and the determination of the relationship between soil respiration and environmental factors such as temperature and water regime are of great importance in understanding the carbon cycling processes in terrestrial and aquatic ecosystems. To evaluate the role that the degraded steppe ecosystems in the temperate China play in global carbon cycle and to determine the factors that regulate soil respiration in these ecosystems, we conducted field experiment to examine the soil respiration rate by using the alkali absorption technique in a degraded steppe community in the Xilin River Basin, Inner Mongolia. We also evaluated the influence of temperature and soil moisture on soil respiration rate. The research site is located in the Baiyinxile Livestock Farm (43°55′ N, 116°19′ E, with an altitude of about 1 200 m). This region has a typical temperate and semi_arid climate. The topography is basically flat with mild relief and the soil is classified as chestnut. The original vegetation was Leymus chinensis steppe, and, due to over_grazing in the past decades, the vegetation has degraded to some extent depending on habitat types and grazing intensity. More than 30 species of plants can be found in the region, among which Achillea frigida，Cleistogenes squarrosa and Carex korshinskyi are the most dominant, followed by L. chinensis, Stipa grandis, Agropyron cristatum, Heteropappus altaicus and Kochia prostrata in terms of their importance value. In addition,Caragana microphylla is sparsely scattered. The maximum coverage is about 40%.The seasonal pattern of CO2 efflux was irregular, though the rate of CO2 evolution was greater in summer than in other seasons. Significant relationships were found between CO2 evolution rate, ambient air temperature and soil temperature (the surface, 5 cm depth, 10 cm depth, 15 cm depth and 20 cm depth, respectively), which could be best described by exponential equations (R2=0.407-0.571 4, p=0.001 8-0.014 1). The influence of temperature was more conspicuous at lower temperature than at higher temperature conditions. This was consistent with the results reported by other researchers. Soil respiration rate was linearly correlated with soil gravimetric water content at 0-10 cm (R2=0.422 5, p=0.011 9) and 10 20 cm (R2=0.500 9, p=0.004 6), but more significant power functions could be obtained after removing the confounded effect by temperature (0-10 cm: R2=0.551 8, p=0.003 9; 10-20 cm: R2=0.645 1, p=0.000 8). The relationship between soil respiration rate (y) and the two variables of air temperature (Ta) and soil moisture at 10-20 cm soil depth (M2) could be described by the following multiple regression equation: y=5 911.648×e0.042 16Ta×M20. 907 58 (R2=0.858 4,p<0.000 1). This equation has much more predicative power than that using temperature and water as single independent variables. The mean soil respiration rate during the study period was 661.35 mg C·m-2·d-1, and the calculated Q10 values based on air temperature and soil temperature at surface, 5 cm, 10 cm, 15 cm and 20 cm depth were 1.63, 1.47, 1.52, 1.70, 1.90 and 1.97, respectively. Both Q10 and soil respiration rate were lower at our study site than at the original L. chinensis community studied by Li et al. (2000) in the adjacent area, possibly due to the difference in water content in the two sites. Our results implied that drought in the growing season tended to have lower Q10 values and lower soil respiration rate. We suggested that the variations in soil respiration and Q10 between degraded and undegraded L. chinensis steppe ecosystems as affected by other environmental factors need to be further studied.