Abstract:For about 30 years the valuation of ecosystem services has been the focus of ecological economics and environmental economics, and China begin the study since later 1990s. This paper firstly in four spatial scales reviews the international study of ecosystem services valuation, as global/regional ecosystem, single ecosystem, watershed, and species and biodiversity. Then the domestic studies up to date are summarized and some notable empirical cases are singled out for their importance. The main characteristics of current study of ecosystem valuation are critically remarked based on four points like study object, value components, valuation method, and spatial and temporal scales. It is showed that, (1) Large scale ecosystems are our common valuation objects, this may be greatly helpful for us to recognize the global or national natural assets, but studies in this scale contribute little to regional ecosystem management. (2)Direct market valuation method is generally chosen for those large scale objects. Unfortunately, the non-use or intrinsic value of ecosystem services are often ignored, so we suggest that the non-market valuation method such as contingent valuation method (CVM), and travel cost method (TCM) should be introduced into domestic studies, only in this way the non-use value and some direct use value such as recreation could be effectively calculated for government environmental decision-making. (3) Almost all the case studies are just in a static perspective, in that the total value is often assessed but the value change caused by land use or climate change is rarely concerned, and we can not establish a clear relationship between one ecosystem and its adjacency, so the in-situ value is better valued than the off-site. Some basic problems in current studies are pointed out, and the key point is that there is not yet a consensus on the understanding of total economic value of ecosystem, so fundamental research about ecosystem services classification and its valuation method is of imperative importance. The most interesting is that valuation results of the same ecosystem vary greatly from one researcher to another, in this sense we hold that the reliability of today′s valuation results might be seriously doubted by the public. What′s more, we find that the valuation results of ecosystem services are seldom used for the government public policy making and this must be greatly strengthened in the future study. Beside this, the domestic must pay much more attention to the theory and method of valuation, because these two aspects still are in a blank state though we have practiced quite many cases. Some other issues for the further direction are also put forward.