Using a Chinese Word Associates Test (WAT-C), this study examined the vocabulary depth of second language learners of Chinese and its contribution to the learners' reading comprehension. Results showed no significant effects of word frequency, word class (i.e., adjectives vs. verbs), and type of association relationships (i.e., paradigmatic vs. syntagmatic) on learners' WAT-C performance. More important, vocabulary depth was found to be a significant and unique predictor of reading comprehension over and above vocabulary size. On the other hand, the relative contributions of vocabulary depth and size depended on what types of texts were read and what comprehension skills were assessed. Specifically, for the long passage comprehension task with questions testing literal comprehension, vocabulary size was a more important predictor, whereas for the short passage comprehension task with questions testing inferencing, vocabulary depth was a more important predictor. These findings were discussed in light of the different levels of lexical complexity between the short and long passages and the different cognitive processing demands the questions of the two comprehension tasks placed on learners.