Abstract:Computer-aided identification for timbers is a fast and accurate method so far developed in wood science. In this paper a system for identifying 500 Chinese hardwoods has been presented, which consists of two programs: WIP (in Chinese) and WIP 1 (in English). This system has many additional functions, such as restooting the previous state, listing the different feature codes among 5 or less tree species as well as modifying information already inputed and adding information for new tree species. The data base consists of 23 multi-characters (Multistate characters) and 19 non-multi-characters(Dichotomous characters).Now there are two types of method dealing with wood identification features: One is Morse and IAWA‘S suggestion which divides the wood features into dichotomous, multistate and quantitative ones, and the other is Weeler‘s opinion which is in accordance with the features of the multiple-entry perforated-card. The method adopted by the authors is in difference from the both methods previously used. Based on the kinds of the features which are divided into 8 groups, such as growth ring, pores, wood ray etc. and then the features in each groups are further divided into dichotomous and multistate ones. The botanical family names are expressed with three letters, i.e. the first letter and first and second consonant letters after that.