This article compares the effects of listening and reading on the incidental acquisition and retention of vocabulary. Two hundred thirty students participated in the study: They either (a) read three academic texts, (b) watched three lectures, or (c) received no input at all and just completed the vocabulary measures. This study also assessed and compared the relationship between acquisition through each of these presentation modes and the following factors: frequency of occurrence, type of word, type of elaboration, and predictability from word form and parts. The reading subjects made greater vocabulary gains than the listening subjects for all four levels of proficiency analyzed. Inspection of pairwise comparisons seemed to indicate that the difference in gains between the reading and listening conditions decreased as the students' proficiency increased. Similar trends emerged for retention. Reading also resulted in greater retention 1 month after the input, except for the highest proficiency students. For this group, no significant difference was found between the listening and reading delayed posttest scores. The relationship among each of the four factors was analyzed and vocabulary acquisition was also found to vary across input modes.