The purpose of this study was to examine the effects of self-disclosure made on the beginning and further the development of close relationship. The proposed hypotheses in the study were as follows: First, modest mode of self-disclosure were expected to produce more liking than the immodest mode of its. Second, negative content were expected to produce more liking than the positive content of self-disclosure. Third, both the mode of self-disclosure and content of its will be existed interaction effect. A 2 (self - disclosure mode) by 2 (its content) factorial design was used in this experiment. The subjects in this study consisted of 200 undergraduates and were randomly assigned to one of the four experimental treatments. Therefore, there were 50 subjects in each experimental group. The data was analysed with the methods of 2-way ANOVA and T-test. The dependent variables was measured by likability scores on 7-point scale. The results were as follows. First, modest mode of self-disclosure produced more liking than immodest of it. Second, there was no significant difference between negative content and positive content of self-disclosure. Third, there was no interaction effect between disclosure content and its mode.